George Miller

Zierler:

Okay, this is David Zierler, Oral Historian for the American Institute of Physics. It is July 13th, 2020. It is my great pleasure to be here with Dr. George Miller. George, thank you so much for joining me today.

Miller:

I'm glad to be able to do it.

Zierler:

To start, would you tell me your title and institutional affiliation?

Miller:

Currently, I'm the Director Emeritus at Lawrence Livermore National Laboratory, Livermore, California.

Zierler:

When did you go Emeritus?

Miller:

I retired as director in 2012.

Zierler:

In your emeritus status, what are some of the ways that you've remained connected to LNLL?

Miller:

I'm available as a mentor for anybody who's interested. Occasionally, the current directors have asked my advice on a variety of things. I've interacted with some of the existing efforts, both in the weapons program, and in the inertial confinement fusion program. The principal activity is I am the chairman of the Science Technology and Transformation panel that advises the commander of the United States Strategic Command. So, the laboratory pays for my expenses to be on that federal advisory committee.

Zierler:

Are there other advisory committees, or are there other ways that you do consulting work?

Miller:

I do a little bit. Up to 2017 when their contract changed, I was on a committee for Sandia Laboratories. I've done a few things at NNSA, for example being on review panels. A principle, current, ongoing activity for me is I'm a member of the Graduate Studies Advisory Board for the College of William and Mary, where I went to school. That's not really associated with the lab, but that's probably the second major activity for me right now.

Zierler:

Right. And you were a three-for at William and Mary, so I'm sure that's very near and dear to your heart, that work.

Miller:

Yes, yes, it is. It's unusual to stay for all three of your degrees at one institution, but it worked out very well for me. I was very pleased. It's a wonderful school.

Zierler:

Well, we'll build up to that. So, to start, George, let's go right back to the beginning. Can you tell me a little bit about your family background? Let's start with your parents. Where are your parents from?

Miller:

My parents and myself were all born in Mississippi. My father got a master's degree in mathematics from the University of Mississippi. He grew up on a farm, but grew up during the Depression, and eventually went and got his undergraduate and master's degree in math. He taught high school for a while, which is where he met my mother. My mother started out as a bookkeeper, and ended up as a bank manager, but then when World War II broke out, he joined the Navy. He spent most of his career in the Navy teaching radar and doing research at the radar lab at MIT. Then, after the war was over, he worked for a while for the Naval Research Lab again in radar, and then in the early '50s, he joined the Central Intelligence Agency. He was responsible for electronic intelligence at the agency until he retired. So, I grew up mostly in northern Virginia. We lived in Germany for three years in the mid '50s, and then I spent one year in high school in Alabama when he was at the Air University in Montgomery. So, we moved around a little bit, but again, most of my early life was in northern Virginia.

Zierler:

To the extent that he was able, did your father involve you in his career?

Miller:

No. I tell people, in many respects, I didn't really know much of what he did until after I got a clearance at the lab and was able to find out a few things. He was very quiet about it. One of the funny stories was when we moved to Germany, he said, "Now, if anybody asks, I'm a civilian working for the Department of the Army." Almost everybody I interacted with were children of military, and they're all very rank conscious. "Is he an officer? What rank? Etc." So, they said, "What's your father?" And I said, "Well, he's a civilian who works for the Department of the Army." And they said, "Do you know where he works?" I happened to know he worked at the IG Farben building on the fourth floor. I said, "You know, he works at the IG Farben building on the fourth floor." And they said, "Oh, he's a spook!"

Zierler:

You weren't fooling anybody.

Miller:

So, yeah, cover stories aside, at least all the kids knew.

Zierler:

Just to foreshadow to your career path, did you have an appreciation for the way that you can integrate a math and science degree into a national security kind of career?

Miller:

Certainly, it was, I guess I would say, implicit. Most of my growing up at the time, my dad encouraged me to stay in science and math. That's what I was most interested in, so that wasn't a hard sell. He actually suggested that I get an engineering degree, and then after I got my engineering degree, go and get a law degree. He said, "The combination of the two of those would serve you very well." Well, I got interested at some point in physics and didn't follow his advice. But again, he was very interested in seeing me stay in the scientific field. Of course, growing up in a national security environment, it wasn't a huge leap for me. I had to decide if I was going to stay in academia, or whatever. Where was I going to work? When I got to it, that turned out not to be a hard decision, because of when I got out of graduate school there weren’t many jobs available in physics per se or in academia. Anyway, it was fairly natural, but as I said before, he did not really talk very much about what he did.

Zierler:

Did you become interested in physics before or after you got to William and Mary?

Miller:

After. I actually went to William and Mary as a chemistry student. So, I was in honors chemistry. Eventually, I got into physics for two reasons. First of all, I hated organic chemistry, because it was all about how you do things. You know, these molecules fit together, and there was a lot of memorization. I happened to have a really, really good modern physics teacher my sophomore year. Physics was more about why things happen the way they are. Between not liking organic chemistry and liking my physics, that sort of pushed me in that direction. My chemistry advisor said, "I'll show you how to get a PhD in chemistry without ever taking another organic chemistry class, but you've got to take the second semester to graduate." And I said, "No, thanks."

Zierler:

The physics department at the College of William and Mary is quite excellent.

Miller:

Yes, and it was actually very early in the growth of the physics department at William and Mary. Again, it's kind of unusual for a small liberal arts school to have a very strong physics department, but around that time, a consortium that was dominated by William and Mary built and ran a cyclotron for NASA. It was actually called the Space Radiation Effects Laboratory. They basically hired away part of the high-energy physics group from Carnegie Mellon. The cyclotron was only 600 MEV, but that was high-energy in the '60s. So, all my professors basically had been at Carnegie Mellon and they came to William and Mary to work on the cyclotron there. The department has grown since then into a very diversified research portfolio. It was starting to do that while I was there, but that was really the founding of the strong physics department there.

Zierler:

At what point during your undergraduate career did you determine that you wanted to pursue a graduate degree in physics?

Miller:

Probably my senior year. I applied for jobs at all of the classic places, many in the Washington D.C. area, NRL, etc., but every summer, I did research at the college. I never went home. So, after my freshman year, I did chemistry research, and then after my sophomore and junior years, I did research in the physics department, and then ended up doing an honors research project my senior year. In doing that, I was interacting with the graduate students a lot, particularly after my junior and senior years. It was just really interesting, so that's really the point at which I decided to go to graduate school. Probably my senior year.

Zierler:

At what point did you determine the kind of physics that you wanted to do? Experimental, theoretical — when did that happen?

Miller:

That was pretty easy. I was never really attracted to theoretical physics. I really enjoyed those summers in the laboratory, building things. One summer, after my sophomore year, we assembled and calibrated a large spectrograph. Then, after my junior year, when I actually started working on something similar to what I did my thesis on, I was doing a lot of electronics and nuclear physics, very, for then, sophisticated timing system to measure very short life times of a radioactive element. I just enjoyed tinkering with things like that. That pretty much settled it for me.

Zierler:

Growing up, were you a tinkerer? Did you like to take things apart and put them back together, that kind of thing?

Miller:

Yeah, yeah. I mean, I was not the classic scientific nerd who did only that stuff, but I was forever taking apart my bicycle and adding new capabilities to it, or cleaning it, and a couple of times when my dad was on a trip we had a Volkswagen Bug, and I took part of the engine apart and put it back together. I was always afraid it wasn't going to work when he came home.

Zierler:

Those were the days when you could take a car apart and put it back together, right?

Miller:

Oh, absolutely. I absolutely could understand how everything worked. So, that was always fun for me.

Zierler:

When it came time for graduate school, did you consider other programs, or did you know you wanted to stay where you were?

Miller:

No, I applied to and got into a number of places. Ultimately, I decided to stay there because I liked the professor, I liked the people, and something that I didn't really — I guess I sort of knew it a little bit, but I didn't really understand how unique William and Mary was until I started interviewing for a job after I got my PhD, and then after I got to the lab, started interviewing people. The thing that was unique was for my PhD thesis, I did absolutely everything. I built the apparatus; actually made some of the detectors that we used; I wrote the computer programs to do the online data acquisition; I analyzed the data. So, I just did everything. I still remember when I was interviewing for a job, somebody asked me, "Well, this is all great, really nice, interesting stuff. What of everything that you told us about did you do yourself?" I was sort of like, "Well, I just did it all." I thought that's the way it was.

Zierler:

Nobody told you otherwise.

Miller:

Yeah, nobody told me otherwise. Again, I think it was also a product of the time when many — I probably was not completely unique, because certainly in the '50s and '60s most university physics departments had a machine of their own. Certainly, from that time on, you get a PhD by doing a little piece of a giant experiment at CERN, or Fermilab, or Brookhaven, or something like that. I saw a little bit of that right toward the end of my graduate student career because the group started doing experiments at Brookhaven. I could see that it's just a much bigger operation, and the ability of a graduate student to take on everything was very limited — time was more precious on the machine, and all those kinds of things influenced what you did as a graduate student. So, I was lucky in that sense.

Zierler:

George, who was your dissertation advisor?

Miller:

His name was Bob Welsh. He was one of the people that came from Carnegie Mellon. There was a group of, I don't know, half dozen professors that sort of all worked together on different kinds of experiments.

Zierler:

What was his area of research? What was he working on at that point?

Miller:

At that point, we were looking at X-rays or gamma rays from meson capture in a variety of materials. There's a variety of things that you can do with mesic X-rays and gamma rays. So, it was really photons produced by capture on a variety of nuclei.

Zierler:

Did you see this dissertation as leading in any particular career path? Were you thinking national security at that point?

Miller:

No. I mean, I guess I would say I was too happy doing what I was doing. I wasn't really thinking about it at that time. I was just enjoying it. When I got out of graduate school and started looking for a job in 1972, there were not that many jobs available in physics. If I remember correctly, I had three post-doc offers. One at Los Alamos at the meson facility – LAMPF, one at UCLA, and one at Carnegie Mellon. And then I had three job offers. One at NRL, one at Livermore, which is where I eventually ended up going, and one at the CIA, where my dad had worked. The people at the CIA said, "We'd like to have you come work for us in foreign intelligence, but it's probably good if you go to Livermore and learn something about what they do there. Then, in a couple years, you can come back." Low and behold, that's what happened, and when they called, I said, "I'm having too much fun. I'm going to stay."

Zierler:

So, what they were offering was, we'll employ you now and we'll send you to Livermore?

Miller:

No, they knew I had a job offer at Livermore, and they just said, "It would be better for everybody if you went to work for Livermore and then came back to work for us after a few years after you learn something."

Zierler:

What was the initial job at Livermore?

Miller:

I guess I was unusual because my dad had worked for the CIA and had clearances his whole life. I came to Livermore with a clearance. So, I started immediately working in one of the nuclear weapon design divisions.

Zierler:

Oh, so you didn't have to start — what is it they call it? The cooler, or the ice box?

Miller:

The cooler.

Zierler:

How did you have a clearance already?

Miller:

They had to do the investigation, but as part of my father's clearance, everything about my history was already well-known. So, they had to do almost no work to check my background.

Zierler:

Right. So, when did you start? Was it the summer of '72?

Miller:

Yeah, August of '72.

Zierler:

And what were your impressions when you first arrived? What did you think about some of the major projects that were going on and how you might fit into the overall objectives?

Miller:

It was really fascinating. One of the things that the particular division that I was in did was as soon as you got your clearance, they basically let you go around and interview with every one of the design groups. There were probably half a dozen different design groups working on everything from strategic weapons to tactical weapons, and really far out advanced designs. So, a wide variety of things, and you interviewed with them, and then, you know, some combination of a number of factors determined my initial assignment — did they need people in that particular group based upon the workload that they had? What were your personal interests and what did the division want in terms of developing people? There was sort of a consensus of those three factors. I ended up working in a group as a member of a team of about four people, working on a strategic weapon concept for the Navy. It was a concept at that time. It wasn't actually a weapon development. So, that was the first design group that I worked on.

Zierler:

Did most of your colleagues come from physics backgrounds?

Miller:

Yeah.

Zierler:

Not engineering backgrounds, mostly physics.

Miller:

Yeah. The way Livermore is organized, it's what we call a hybrid matrix. Most of the people live in discipline oriented divisions, sort of like a university. That resulted from the fact that for a very long time, the lab was run by a university, University of California. One of the principle founders was E. O. Lawrence, a Nobel Prize winner and in many aspects the founder of “big” science. The organization was based on his ideas. So, there were engineering divisions, a physics division, a chemistry and material science division, and computational divisions, all of the basic disciplines. But there were a small number of what were called programmatic divisions. The weapons design divisions, there were two of them, were basically all physicists, but they were basically in charge of designing nuclear weapons. They had engineers, chemistry and material scientists and computational experts that in a programmatic sense worked for them to accomplish the national security tasks at hand. So, I went into one of those programmatic divisions. We used to say that people throughout the lab thought we were at the top of the pyramid, and we thought we were at the bottom of an inverted pyramid.

Zierler:

George, in what ways, if at all, did your dissertation prepare you for this kind of work? In other words, how did you rely on your academic appreciation of physics in your day to day at Livermore?

Miller:

I guess I would say in two ways. In many respects, getting a PhD is a giant test to see if you can learn, and learn pretty much on your own. You know, it's different than an undergraduate. You do a lot of stuff on your own, and you're exposed to a very broad segment of physics. Atomic physics, nuclear physics, plasma physics, relativistic astrophysics, just a whole broad array of things. So to start with, having a cursory knowledge, maybe more than a cursory knowledge, but a knowledge of that very broad segment of physics. Every one of those has something to do with a nuclear weapon, so being able to access all of that knowledge as a starting place. A starting place because I needed to know a lot more detail than I ever did in an academic class on many, many different aspects of those things. The second, particularly as an experimentalist, was that I learned to organize my work. I learned to organize the work of other people, because some of the stuff that had to be built was built in a machine shop, and you'd have to figure out how to tell the people in the machine shop what you wanted done. There were usually younger graduate students that were helping you, so there was a lot of, organizational and teamwork that was associated with my degree as an experimentalist. So, both of those elements, I think, helped me a lot in many, many different ways in the early part of my career.

Zierler:

For that initial design group that you were a part of, what were some of its objectives?

Miller:

This particular design group was developing a proposal for the warhead that eventually became the trident warhead missile system for the Navy. That assignment was actually given to Los Alamos, but there was always a competition between the two labs, and this was Livermore's proposal. But then the government evaluated our proposal and the Los Alamos proposal, and they picked one. So, it had a very long list of what we would call military characteristics: size, weight, yield and safety. So, a very long list of characteristics, and our job was to try to come up with the optimum design that met all of those characteristics. That's generally speaking what all of the weapon development activities in those two design divisions were principally about.

Zierler:

I'm interested in this idea of an inverted pyramid versus a regular pyramid, and where you were in the policy process. Probably, in that first job, you didn't have such a good appreciation of that, but as you rose up at the lab, obviously, you knew a lot more about these things work. Can you talk a little bit about the general production and information flow? In other words, if you're charged with the particular task of coming up with specifications for a warhead, where does something like that initiate, and how does it get to the actual design group?

Miller:

Well, the characteristics themselves are generated by the military. So, they have a certain military mission that they want to accomplish, whether it's a bomb, or a reentry vehicle, they have a certain set of characteristics or capabilities that they are interested in. So, basically, you start with that set of characteristics. When I was doing design work at the lab, this was the early '70s, there was roughly speaking 20 years of history. So, there were a lot of things people had tried. There was the historical knowledge, and then there's your own creativity. So, that's where it starts. But then you very quickly have to start interacting with what I'll call the real world. Whatever your idea is, it's got to be built. So, there are machine tools that have to create the pieces and have to be created out of real materials. So, you'll very quickly engage the engineers and the material scientists. Probably the other two biggest elements were the modeling and simulation, so computers, and testing. In those days, the models were far less sophisticated than we knew was required to adequately model what goes on in a nuclear weapon. We all understood the very fundamentals of atomic physics that was going on or the nuclear physics, and there's no way a CDC 7600, or 6600 could simulate them. Those machines are more than a million times smaller than the computer that you have on your phone today. There's no way that it could accurately simulate all that, but it was such a complicated nonlinear problem that you had to use simulations to guide you. Those simulations had to be grounded in data, and that data came not only from nuclear tests, but you got a lot of data from other places also. So, it became very clear, very quickly, that I had to understand the details of the models. Already, by the '70s, they were more sophisticated than I as a designer had the time to create them myself. So, there was already beginning to be a bifurcation in terms of design physicists who used the codes, and other code physicists built them. In the early days many of the designers actually built the codes that they used for design. As a design physicist, I had to understand all of the approximations that went into the code so I knew where it was good and where it was not good, and then design an experiment to tell me all of the things that I knew I didn't understand; get a normalization for the simulation. So, all of that together, in some sense, is why I felt the bottom of a giant pyramid. I was feeding all of this information up, and the weight of everything that everybody was doing was based on what I told them to do.

Zierler:

What was the deliverable for you? Was its analysis, or was there an actual finished product, or a prototype?

Miller:

There was a finished product. If it was a concept, it was a nuclear test in the desert that you exploded. So, as you learn very quickly, nuclear tests are very complicated, and they're very expensive, and there are not very many of them. In many respects, your career was dependent upon how that thing that you designed and directed behaved. So, you very quickly were given a tremendous amount of responsibility.

Zierler:

Did you attend nuclear test demonstrations?

Miller:

Yeah, every time it was one of mine, I was there, with my hand on the wall, waiting the 15 or 20 seconds for the seismic shock to come from the test to the building that I was in. I claimed my hand was calibrated, and I could tell whether it was okay by how much the building shook.

Zierler:

Did you have any duds, or the all worked out?

Miller:

That's an interesting story. The first test that I worked on, I was a member of a group. So, it wasn't my design, it wasn't my concept; I was learning as part of a group by doing a particular kind of calculations. I can't remember the exact timing but let me say I had been at the lab 15 months, so not very long, and they gave me an experiment of my own. It was a really different, brand new kind of a concept, and it did not work. So, the first thing that I was responsible for on my own didn't work. There were two things that happened. The first is I probably learned more about nuclear design trying to figure out why it didn't work than I ever did on anything else. Not only for that design, but about the whole issue of how you design, and what you worry about. One of my mentors, a mentor to lots and lots of people at the lab, Johnny Foster, said, "As a designer, you have to learn to turn over every rock." That set of lessons, skepticism about what you know and what you don't know, just the whole set of views of the world, were formed for me in that particular experiment. The second thing, which again is kind of interesting, is that the lab at that time was willing to take risks. This was a very difficult design, and in the end, I turned up some new things. In many respects, the reason it failed was not foreseeable, at least not by me, and it was okay to take a risk for the right reason, and fail, because you learn. You have to pay attention to a lot this in the broader context. When he was the commander at STRATCOM, John Hyten used the North Koreans as an example of how to go faster. Try something, it fails, try something again. This very fast, iterative cycle, that's what the country did for decades, and that's how we made progress. We've become so risk averse. You'll study something for 15 years before you'll ever cut metal, and then there's something you always forget, and that's why it takes so long. But I grew up at a time at the lab when taking risks, not only was it okay, it was encouraged. One of the first pieces of classified material I was told to read as I came on at the lab was written by one of my predecessor directors, John Nuckllos, and it was called “A Baker's Dozen of Failures”. He analyzed failures in our own test history as a way of learning how to think about this whole issue. So, the first shot that I was responsible for didn't work, but the next dozen or so that I did, did work. That first test was a very important learning experience —

Zierler:

It was okay to fail. There was a culture of acceptability that failing was productive as well.

Miller:

Yeah. If you failed for the right reasons, and you learn something from it — again, this idea of you're not moving forward if you're not taking some risks. But they need to be taken in an intelligent fashion.

Zierler:

As you said, this was a culture that was encouraged right from the top.

Miller:

Yes, absolutely. I mean, it was implicit and explicit — many of us have remarked that the lab had changed since its inception, but it was still Lawrence and Teller's lab. The culture of the lab was dominated by those two people. The culture of big science and big teams came from Lawrence, and the culture of trying new things came from Teller.

Zierler:

Did you get a chance to meet Teller or work with him?

Miller:

I wouldn't say I worked with him. I certainly interacted with him. When I was director, one of the comments was that the good news and bad news about being director is all the previous directors are still alive. The good news is, you're going to live for a long time. The bad news is they'll continue to tell you what to do.

Zierler:

George, just to fast forward a little bit, in terms of the culture of risk aversion taking over, to what extent was that about a tighter budgetary environment? In other words, in the early days, was there more room to make mistakes because there was more money to make mistakes?

Miller:

I would say more “more money” in two respects. I don't mean more money in the sense of lots more in terms of actual dollars—first because the administrative overhead was significantly less. So, the fraction of dollars that went into real work versus the fraction that went into administrative tasks was very different than it is today. Secondly, there were far fewer “bins” for the funds we managed. We only had two – R&D (research and development) and T (testing). The government told the system WHAT to do and the Labs and plants figured out HOW to do it. It was a different time and a different culture. It didn't matter whether you started at the Atomic Energy Commission, or you started at the director of the lab, or you started at your division leader, there was a significant culture of, push the detailed decisions down as far as you can. Throughout that whole system, taking risks were okay, and they were encouraged. If you look up the history of the word bureaucracy, it's a French world, and bureaucracies were built to keep the king from doing something. So, as bureaucracies got built within the system for a variety of reasons, and people started adopting a culture of punishing mistakes, people just got really conservative. I mean, it doesn't matter whether you're talking about the culture at the lab, or you're talking about military aircraft. Go look between World War II and 1970, I think there were sixty different kinds of aircraft built. Every few years, a new kind of aircraft. And now you've got an F-35 that takes 30 years. All of that is associated with trying to get it exactly right, and doing studies, and analysis of alternatives. All of that engenders a tremendous amount of conservatism. That's what General Hyten was talking about. He's not the only one. We need to change the culture. This is a significant part of many of the recommendations that are going around the defense department. Look at the way the commercial world operates and look at the way the government operates. They do things on very different timescales. The commercial world does things on 18 month timescales, not 18 year timescales. They're constantly turning things over, learning and improving. In many respects, that's the way the national security culture, really up through the '80s, in this country, ran. What we're finding now is that many parts of our adversaries’ systems are operating on a much faster time scale than we are. — I would never trade our culture for that of any of our adversaries, but they're operating differently than we're operating right now. That's a problem.

Zierler:

You mean, it actually has national security implications?

Miller:

Absolutely, it does.

Zierler:

George, what was your next job after your initial work at the design group? At what point did you get the sense that you were on a track towards leadership?

Miller:

Let's see. The next job after running a design group was, I was the project leader for the development of the warhead for the ground launch cruise missile. That's not just preparing for a nuclear test. That's actually going through all the steps to put something into production. So, it involves much more extensive interactions with the production complex, etc. So, I was the first project leader for that program within the Laboratory. At Livermore, there usually is a project leader who's a physicist, and a project engineer. So, there's really a dual responsibility. That's about the time that you started having extensive interactions outside the laboratory, both with the uniformed military, with what was then the Atomic Energy Commission, and the production plants. So the horizons started broadening about that time in my career. About that same time, the lab started letting me take some courses outside the lab. I spent two weeks at Hudson Institute, and it was absolutely amazing, because at that time, Herman Kahn was the director of Hudson Institute. I got to spend two weeks with Herman Kahn and people like Bill Schneider who was the head of OMB, and head of the defense science board. So, it really broadened your view of national security, how the system operates and the interaction of technology and policy. So, that would have been about 1978. I'd only been at the lab six years by then, but in six years I think I had done something over a dozen nuclear tests.

Zierler:

At this promotion, was this the first time that you got a sense of where national security policy interacted with Livermore?

Miller:

No. I guess I would say my third design assignment. I was put in charge of the group, half a dozen people, that was designing part of what became the B-83, the large lay-down bomb. It was actually called the B-77 at the time, but it became the B-83. That involved, more interactions with the military, and with the production complex because it was a weapon for the stockpile, and because it was going to get produced. Interaction with policy started fairly early in my career, but there was a significant difference when you're in charge of part of the lab's interactions as opposed to just being a participant. But even in the early days, the lab has always had a broad engagement with the national security community. I remember very early in my career, there was a lecture given by General Russ Dougherty, who was the four star Air Force general in charge of the Strategic Air Command. He came and gave a talk at the lab. Admiral Rickover came and gave a talk at the lab. I remember Sid Drell coming and giving a talk at the lab. This was like — I can't believe this. I took a class from his book, and here I am listening to the man. We actually became very well acquainted later on. So, the lab invited, regularly, national leaders to come in and give talks to the broad segment of the laboratory. It was a broad spectrum of views — General Dougherty and Sid Drell both strongly supported national security, but they had a very different perspective on what's important. So, it wasn't just a singular view. It was a much broader interaction.

Zierler:

George, at what point did you stop operating as a physicist on a day-to-day level? When did that happen?

Miller:

Let's see. Literally, as a working physicist, probably by the time I became the project leader for the ground launch cruise missile. Although, when I became a division leader in 1980, and deputy associate director in '84, and associate director in '85, there were still many decisions that I had to make that were based on my understanding of physics and what was going on in a nuclear device at a very detailed, fundamental level. I still remember going in very early in the morning before anybody came, or staying very late at night, trying to work through some physics issues to make sure that my decisions were on sound basis. But I wasn't literally doing physics after I started getting promoted. That had sort of changed for the lab about the time I came to the lab. When I came to the lab in '72, Roger Batzel had just become director, and Mike May had just stepped down as director. I think he stepped down in 1971, and Roger became director in '72. Anyway, when Mike May was director of the lab, he wrote a seminal paper on computational astrophysics. So, people's ability to do technical work while managing was beginning to change fairly rapidly in the early '80s. I frequently tell people when I became an associate director, which is the level right below the director, in charge of major segments of the laboratory, I probably spent 80% of my time worrying about what was going on inside the laboratory, making programmatic decisions, doing reviews, but it was basically inwardly focused. Five years later, I was spending only 20% of my time as an associate director inside the lab. And this is part of the whole bureaucracy conversation. Probably 80% of my time was spent interacting with various communities in Washington, explaining what the Lab was working on, discussing external factors, dealing with increasing external regulation of environment safety and health. All kinds of things besides the detailed, internal programmatics.

Zierler:

In what ways did your responsibilities change when you became a division leader in 1980?

Miller:

When I became division leader, I was responsible for all of the design work and all of the computational physics work, and some of the basic physics research in the thermonuclear design division. Basically, my responsibilities were reviewing all the projects, so making sure that they were being done on the highest quality technical basis and making major decisions about what projects for the division to pursue, and setting future direction. Decisions about how to do it, and decisions about what to do in a broader context. So, as an example, by 1980, it was clear to me that at some point in the future, there would be more restrictions on nuclear testing. There already had been a couple of major ones. The one that I personally participated in was the 150 kiloton test limit. It became clear that there were going to be, at some point in the future, further restrictions on nuclear testing. So, I did a couple of things. I started a couple of programs to substantially enhance our understanding of some of the basic physics of nuclear weapons. A lot of the physics was in the models, but not in detail, We didn't actually have experimental data or, in many cases, not even theoretical basis upon which to understand the complicated physics going on. So, I started a couple of programs to better understand the basic physics of a nuclear device. This eventually became the basis of the Stockpile Stewardship Program; we were already thinking about it that far ahead - the stewardship program wasn't for another 15 years and we were already beginning to think about it. Again, with University of California culture, one of our responsibilities was to be prepared for whatever might happen and do whatever the government might want to do. In a similar vein, because of my personal experience, it was clear that there were many things about the physics of nuclear design that you could not learn in an underground test. The underground environment is too harsh. Many of the things that you want to know occur deep inside the device, and you actually can't even get access to them experimentally. So, we started some programs particularly associated with getting data on the then very, very early laser capability that was getting started. So, I started some new theoretical programs, and started some new experimental programs, By the time I got to be the deputy associate director in '84, I insisted that the lab develop programs to begin using parallel processing computers. All the mainline computing at that time was serial. Everybody thought parallel computing was just too hard. Many people thought that no one will ever figure out how to use it. The Lab purchased some of the very early parallel computers, and I basically said, "Okay, here's some money, and the only thing you can do with it is research how to use these computers. You can't use it for anything else," because very few of the main line code efforts wanted to do it. As a division leader and a deputy associate director, those are the kinds of things that you try to do – set the direction for the future. You worry about not only the quality and the specifics of what the current program is, but to really start thinking about what things are missing.

Zierler:

How broadly do you have to conceive in terms of considering what's missing? How far out do you have to look?

Miller:

It's always a complicated trade-off between how far you want to look, what confidence you have at predicting the future, and what you think is actually possible. I mean, it was a huge bet at the time to put a little bit of money into laser experiments. The lasers, the diagnostics, everything was really crude in the mid '80s compared to today. It's taken us really until NIF came up that we really had the ability to start measuring some of the things that we cared about. But we did some interesting experiments, even, in the mid '80s. I can't remember exactly when it was, but we did an experiment measuring the opacity of partially stripped iron. It was recognized as the most significant astrophysical measurement of the year. So, we made an astrophysical measurement in the laboratory. That’s what's so interesting about being a technical leader. – if it’s the right kind of institution. The Laboratory senior leaders are really technical leaders, not administrators. You get to make important technical decisions about the future direction.

Zierler:

It does sound like in certain opportunities, there is a culture of pursuing basic science.

Miller:

Absolutely. It's always basic science, but it's for the purpose of advancing a program. But in these kinds of programs, there's lots of basic science that you need to know in order to do a better job on your application. So, we're not just doing research for research's sake, but you're doing research for the purpose of advancing a particular program goal. But if you do that right, the labs have done absolutely world class stuff while working on applied science.

Zierler:

George, another national policy question. As you've risen in leadership, and when someone like Ronald Reagan comes in with his ideas on national policy and nuclear weapons, how quickly until you feel those changes at the lab? In other words, how quickly is there the translation from the Oval Office to the military to the labs, and how do you feel those changes?

Miller:

It's different for different Presidents, and it's different depending upon exactly what you’re talking about. Certainly, we transitioned from doing nuclear tests to not doing nuclear tests, fundamentally, in a few months as a result of Congressional action and acquiescence of President Bush. In the case of Reagan, we started feeling that right away also. Within a few months, they formed the Strategic Defense Initiative that General Abramson led. I was at a meeting at Cape Kennedy with him in charge, within a few months. I think Reagan's speech was in March and by that summer, they had brought together maybe 50 scientists and technologists from around the country to try to lay out the beginnings of the SDI technical program. In the case of Livermore, it's probably a little unusual because of Teller, and his relationship with the administration. Although I think his influence on SDI has been overblown, he certainly had influence. So, we felt changes pretty rapidly, certainly within a few months. Other times, it's really slow, particularly today. Because of the nature of the authorization and appropriations committees in Congress, it can take years, because you've got to go through the whole Congressional cycle before you can get funding allocated. When I became the associate director in 1985, we had two budget categories. One was called R&D (research and development), and the other was T (testing). The government told us, "We want you to do the following weapon production, or weapon development programs, and the military said we're interested in solving this problem or that problem.” With that guidance, the Labs planned their programs. Apart from supporting the weapon development programs, the lab had complete flexibility with what it was doing with all the rest of the money. How many nuclear tests we did, that was our discretion. Whether we put it on applied physics, basic physics, how much did we put in computational physics, how much did we put in lab experiments? Those were all decisions that were made at the lab itself. Our job was to support specific weapon development and sustainment programs, to be prepared for future military needs and changes to the threat environment, understand the basic science and engineering of nuclear weapons and push the boundaries. How we did it was our responsibility. I think, today, there are 123 budget categories. The lab doesn’t have the authority to move money from one category to another to support changes in priorities. The ability of the lab to move money from one place to the other is highly constrained. There is more flexibility with the LDRD Program (Laboratory Directed Research and Development) but much less than was true historically within the weapons program itself. That's another reason why things are very slow. When we were doing a weapon development program in the past, if we had a problem, we just internally moved money and people from an existing program over to the weapon development program, because that was higher priority. Our bottom line budget for decades was constant. So, we had basically level of effort funding, and what the government did was it said, "Here are our priorities." So, the government decided what you do, and the priority of what you do, but the lab figured out internally how to balance across all of those competing desires. It led to just much more like a company today, where they can make internal decisions about how to move money around to support the highest priorities. So, basically, through the '90s, all of that went away.

Zierler:

George, what were some of the technical challenges and opportunities posed by SDI, for you personally, and for the lab as a whole?

Miller:

If I could trivialize it, on a technical side, it was just a very hard problem — I mean, the lab just loves terribly hard problems. It just excites you. That's what gets your technical and scientific juices flowing, to have a really hard problem, and SDI was our really hard problem.

Zierler:

Was the feasibility of it from a physics perspective ever in doubt to you? Did you ever wonder how is this actually going to happen? Or did you think that it was possible from the beginning?

Miller:

Let me answer that two ways. I'll use the X-ray laser as an example. Did I think we would make an X-ray laser? Absolutely. It was a really hard problem, but I believed that if we worked at it long enough, we would figure out how to make an X-ray laser. Did I think that it would be a useful military object? Many of the people in the lab, myself included, were very skeptical.

Zierler:

Did that matter? Did anybody care about your opinions on the military utility of such a thing?

Miller:

Oh, boy, did they ever.

Zierler:

So, it wasn't just like, "We don't care what you have to say. You're the physicist. Just make this thing happen." They actually cared about your perspective on military feasibility.

Miller:

Oh, yes, and that ultimately is how I became an associate director. My predecessor, Roy Woodruff, who was a very close personal friend, resigned because he believed that Teller and Lowell Wood had far too much influence in terms of the military utility of the work that the lab was doing, and that the director wasn't keeping them under control, or allowing alternate points of view to have access to the same levels of government. Because of who Edward was, he had access to levels that Roy didn't. Roy was certainly prepared and able to make his own arguments about what the feasibility, but because of Teller's access he believed that the government wasn’t receiving a balanced view of the system’s capability. That was important, because of the University of California connection, the lab has always not only tolerated but encouraged wide diversity of thought on any subject. The focus was always technical, so we would have incredibly difficult and confrontational arguments about technical subjects. But in this case, it also boiled over into political ones. So, the fact that the director allowed Edward and Lowell to be very optimistic, and Roy to be much more cautious, is not unusual. You'd have to ask Roy his personal view. My own view from the outside was that the issue was that Edward and Lowell had access that Roy didn't, and that people at high levels were getting a biased view of things. In my own personal example where SDI is concerned, I was responsible for a set of SDI programs that were probably worth $350 million a year, a very substantial fraction of the lab. I was asked by Bennett Johnson, who was the senior senator on energy and water from Louisiana, to testify on SDI. In particular there was a substantial debate on what General Abramson called early deployment. They wanted to begin deploying some of the technologies very, very early. I was asked to come testify, and I got calls from Abramson, from two personal friends who were generals, my program leader in Washington, saying, "You have to support the administration position." And I said, "I'm sorry, I work for the University of California. I strongly support doing the research, but I don't think we know enough to begin early deployment." That's what I said. People care not so much about our political views, and the lab tries very hard to avoid a political statement. But the political implications of a technical statement are often very important. So we frequently get asked about our technical views. More recently, certainly in the '90s, there was a lot of discussion, including testimony and a lot of behind the scenes discussion about the comprehensive test ban, and the whole set of issues associated with Stockpile Stewardship. An example that's different than SDI is, again, a personal example. During one of the previous interests in a comprehensive test ban, some senators asked a particular person in the lab, Ray Kidder, whose views were well-known, to write a report on why a comprehensive test ban was okay. The director said, "Absolutely. Ray should write whatever he wants. But he asked me to do a complimentary evaluation. This was in around '86, or '87. He asked me to discuss my views on the issues associated with the comprehensive test ban. So, the lab has almost always had internal, and frequently it becomes external, this dynamic of allowing different points of view. As long as you pretty much stick to the technical stuff and the technical quality is there, it's not only tolerated, but it's encouraged, even when it generates controversy. There are many, many examples of this in the lab's history.

Zierler:

You mentioned that part of your work on SDI helped lead to your being named Deputy Associate Director in 1984. What were the broader issues at play there that led to this promotion?

Miller:

Actually, SDI led me to become the associate director because my friend resigned. I was appointed as the associate director, after his resignation. When I became the deputy associate director, that was just a retirement of the then head of the weapons program at the lab. Roy became the associate director, and then he promoted me to be the deputy associate director. So, it was sort of a reorganization based upon somebody's retirement.

Zierler:

What were some of your key responsibilities in this position?

Miller:

As opposed to when I was a division leader, when I had a broad set of responsibilities for thermonuclear design, I was then responsible for the entire nuclear design program. With becoming the deputy associate director, I was responsible for the fission design division, the thermonuclear design division, and a division in the physics department that was responsible for developing and maintaining a lot of our physics databases, and a lot of the physics research that was done. So, basically, I was in charge of all of the physics research associated with nuclear design, and computational physics. When I became the AD, then I had responsibility for the weapon development programs, all of the interactions with the military on early interests, so it basically included the entire weapons program at that point.

Zierler:

George, can you talk a little bit about the role of computers in weapons design, and how that changed over the course of your tenure?

Miller:

Fundamentally, it didn't change. What changed is that as the computers got bigger and more powerful, the level of physics that you could simulate became more robust and more detailed. Certainly, by the time I came to the lab in '72, they were just making the move from codes that were primarily one-dimensional. So, when I say one-dimensional, that is, you basically assumed everything was either a sphere or a cylinder. The size of the computation goes up as the number of dimensions — if it's one-dimensional, all you care about is R. If it's two-dimension, it's R squared, and if it's three-dimension, it's R cubed. So, the size of the problem exponentiates very, very quickly with dimensionality. So, in 1972, they were just beginning to make the change from one-dimension to two-dimensions. As you can simulate more dimensions, that gives you a little bit more freedom as a designer. So, again, more recently, the move to three-dimensions is a big deal because it allows you to simulate the real world more accurately. For many problems and many details, you actually can't get a particularly good answer if you just assume two-dimensional symmetry. While many different types of simulations are used, it’s often helpful to think of Lagrangian codes. The codes basically discretize the real world into chunks. So, if it's one-dimensional, the size of the computer might allow you to take 1000 chunks to describe the physical entity. If it takes 1000 zones to describe all of the physical processes in one-dimension, then in two-dimensions it's 1000 squared, and in three-dimensions it's 1000 cubed. So, you can see the problem size goes up drastically as you increase. So, those are the set of changes that take place. So, there's dimensionality, and then there's the amount of physics that you can simulate. As computers get bigger, you can also add more physics and you can describe the physics that’s being simulated with more detailed models. It’s always important to remember that in the end, they are models – no matter how detailed. Fundamentally, the comment is, "All simulations are wrong. Some are useful." That's very much the case in a nuclear weapon. The physics of a nuclear weapon is very complicated, and requires a coupled set of nonlinear set of equations. It's just manifestly really complicated. So, the trick is to figure out how to use the computer code to point you in the right direction. I frequently likened it to you're trying to survey the top of a mesa, and you want to end up in a place that's very, very safe, and you don't want to fall off the edge. So, you try to use the computer code to tell you where the edges are that you might fall off of, and what you have to do to stay away from them. That's, in many respects, how the computer simulations were, and still are, used. None of them are or will ever be “first principles” in anything that resembles my lifetime, or our lifetime, or my children's or grandchildren's lifetime. The number of phenomena that you have to worry about and the timescales over which things happen make “first principles” calculations a long way off — things happen on the femtosecond time scale, and happen on the microsecond timescale. So, you have to cover all of those. For example, just the processes that go in partially stripped, high Z atoms, can't be simulated in full detail today. So, everything in the codes is a model and the adequacy of the model is often determined by the computer used for the simulation. But since they are models, then need to be normalized by data. Even for example, the physical relationships between pressure and temperature. People think it's PV=NRT. Well, that's a really simple formula for the relationship between pressure and volume and temperature. The real world is much more complicated. That's an ideal guess. So, even those fundamental relationships of how the materials behave has to be simulated. All of these codes have normalizations. The normalization data comes from many sources - including, but not exclusively so, from nuclear tests. So, again, what you did as a designer was you tried to make sure that you have code that was reasonably normalized, so that you sort of knew where the edges were. Then you change the design get as far away from the edges as possible within the constraints. A concept that we used, I guess I would say, in an implicit fashion — it was not as explicit as it is today when it was really popularized in the days after stewardship by Mike Anastasio — is just called margins and uncertainties. If you know that a thing is going to fail in a certain area, you want to make sure that your worst operating point is pretty far away from where you know it fails. But your knowledge of each of those is uncertain. You want there to be space between the worst operating point with its uncertainty and the failure point with its uncertainty. We used this concept all the time, even though it wasn't formalized quite the way it is today. That's the idea. We don't exactly know where the edges are, and we don't exactly know where on the top of the mesa we are, so we want to be pretty far away. That's how you use the codes that are normalized by the data, both from nuclear tests and from other sources, to help guide your decisions ensuring that you have enough margin.

Zierler:

George, when did you get involved in the stockpile stewardship and management plan, and what were some of your primary contributions in that initiative?

Miller:

In many respects, the thoughts surrounding Stockpile Stewardship were around a long time. As I said, back in the '80s when I was a division leader, I started working on things like this, and I, and many other people, started programs to understand the basic physics better. Those ideas were part of the culture that I grew up in within the Laboratory; the ideas were explicit and created by the people who formed the Laboratory. At the end of President Bush’s term of office, the Hatfiled – Exon – Mitchell Amendment to FY 1993 Energy and Water Appropriation established a moratorium on nuclear testing through July 1, 1993 and directed negotiations on a Comprehensive Test ban to be concluded by September 1996 – with 15 nuclear tests being permitted during the intervening period. During the Hatfiled-Exon-Mitchel debate, in the spring of 1992, prior to the election of President Clinton, a small group of us, Bruce Tarter, who was Deputy Director at the time, myself, Wayne Shotts, Larry Woodruff, and Bill Lokke, got together for a week and said, "Okay, when the new president comes into office and wants to talk about more restrictions on nuclear testing, what should we tell him from a technical point of view? Over the course of a week, at a conceptual level, I think we created the fundamental ideas of the Stockpile Stewardship Program. When President Clinton got elected, and Secretary Hazel O'Leary named Vic Reis to become the Assistant Secretary for Defense Programs, Larry Woodruff and I went and gave a briefing on what we thought about these issues before he was officially confirmed in the new job. Vic is an absolutely fantastic, wonderful person. I mean, he is a genius in many respects. He took our ideas and made them his own. He worked with the Defense Department, particularly General Shalikashvili and turned those crude ideas into what became the Stockpile Stewardship Program. After Vic became the Assistant Secretary, he had a group that was called the Navigators, which were basically his senior advisors. None of us were in the department, so from each lab, he had two people. He used to say, "I want a bomber, and I want a physicist," not realizing that he was pissing me off by saying that. He thought I was a bomber, and not a physicist – you can be both. So, it was me and Dick Fortner from Livermore, John Immele and John Browne from Los Alamos — John Browne eventually became director of Los Alamos — Roger Hagengruber and Jerry Yoenas from Sandia, and Steve Gedice who represented the production complex. He met with us probably three or four days at a time, and together we crafted what became the Stockpile Stewardship Program — all of the outlines. Later on, DOE created all formal documents and management plans. This initial effort occurred about the same time as the President was trying to decide what he wanted to do for his nuclear test initiative. I think the general consensus below the cabinet level of the Defense Department, the State Department, the Department of Energy and the nonproliferation and arms control community, was that the next step should be a limit of about 1 kiloton, following up on the 150 kiloton limit. Many senior advisors, including people like Harold Brown agreed. The broad view was that you could likely have very high confidence in your stockpile if you can still test at a kiloton, and at a kiloton, the uncertainties in verification are probably small enough that you don't have to worry about it. So, most of the community was talking about a threshold. About that same time, there was a conference in Omaha that was chaired by the commander of STRATCOM, Admiral Hank Chiles, trying to come up with a view of what the Defense Department should support. So, all of these things were sort of happening at the same time in my memory. I was one of the major participants, along with one of Admiral Chiles's deputies, to orchestrate and script that conference. People call it a confidence conference. I gave one of the major talks and was one of the major proponents of what is now the annual assessment process. In that process, the lab directors write a letter every year, and it goes unfiltered to the president, and then the commander of STRATCOM writes a letter on the military effect of the lack of testing. So, I was certainly right in the middle of almost every aspect of that. Another major participant, if you haven't talked to him, is Mike Anastasio, who was the director of Livermore and Los Alamos. He personally gave the briefing to General Shalikashvili. I was supposed to, but I was in Hawaii when it finally got scheduled, and I didn't want to come home from Hawaii, so Mike gave that briefing. I guess the simple answer, I was in the middle of many significant aspects of the formation of the Stewardship Program.

Zierler:

George, if we can zoom out a little bit, what was your sense of the overall impact of the end of the Cold War on the laboratory? First, was it a gradual change, or did it sort of happen almost as fast as the collapse of the Soviet Union itself?

Miller:

I would say it happened very rapidly. The details of when everything happened, stretched over several years. It took a while from when we stopped testing before Stewardship actually was in place. But basically, everything happened pretty much all at once. You had the Hatfield-Exon-Mitchell bill, which stopped nuclear testing for a period. According to the bill, we could have gone back and done a nuclear test, and eventually the decision was made not to. Most of the weapon development stopped at the same time. So, you had the stopping of nuclear testing; you had a significant hiatus in nuclear weapon development; production went almost to zero. You had all of this stuff happening simultaneously, and it was incredibly traumatic. Also, about the same time, while Hazel O'Leary was Secretary of Energy, we had the Galvin Commission, which basically said we ought to close Livermore. That recommendation wasn't adopted, but just recommending it had a major traumatic effect. And not too long after that, 9/11 occurred, the suggestion was made to move Livermore to Homeland Security and have a major shift in National Security efforts to focus on counter terrorism and non-proliferation. So, through that decade of time, a lot of stuff went on. I would say, again, in simple answer to your question, particularly the weapons program at the lab had to fundamentally redo everything at that time. The way in which we trained people was through nuclear tests and weapons development. We had none. The majority of the weapons program was focused on doing weapon development or nuclear tests. So, what do you do when you come to work every day? It's not only the what you do, but the cultural organization. How things are organized was turned upside down. Again, about this time Hazel O’Leary, the Secretary of Energy, came out and gave a talk at the lab in which she basically said, "I know most of the people in this room think that you're going to have a lifetime career at Livermore, but everybody should change jobs every five years." For many people, it takes five years before you really can understand a nuclear weapon. So, again, all of these things are happening simultaneously. I think it's a tremendous credit to the laboratory that we were able to cope with these monumental changes and not only survive but thrive. Certainly, many, many directors played a part in getting us through this. But, in my view, it was a tremendous accomplishment for the laboratory to come through such a traumatic period, and basically retained its fundamental culture, and retained its commitment to excellence. In a very fundamental way, it was Stockpile Stewardship and Vic Reis that gave us the required focus. Things are very different now. The Lab does things in a very different way and it's just a tremendous accomplishment to have been able to make that happen.

Zierler:

It's interesting, because it sounds almost as if the nuclear weapons testing program was so wrapped up with the Cold War and containing and deterring the Soviet Union that absent that challenge, there wouldn't be any need, or very little need for a nuclear weapons component. But, of course, we still have one today. So, what were some of those early decisions about the future of nuclear weapons testing and design, absent of Soviet threat?

Miller:

It pretty much all happened at the same time, and as I said, the first element that was so important about Stockpile Stewardship was that it gave the labs — Livermore, Los Alamos, and Sandia — gave them a focus. The focus was to significantly advance computing capability; significantly advance the experimental basis upon which you can collect data about nuclear weapon; significantly advance our understanding of the fundamental physics that goes on in a nuclear weapon; and figure out how to maintain a technological object, the longevity data for which rarely goes over 15 years because that's how often they were turning over, and make it last for 40 years. It's sort of like SDI. Vic and Stockpile Stewardship gave the labs an incredibly complicated and difficult technical challenge to sink their teeth into. That's what we did. We turned to all of that. When Vic and the labs conceived of Stockpile Stewardship in the early to mid '90s, we thought that was a program of work for 15 or 20 years. But you can't just study and analyze forever. This is a comment that both Mike Anastasio and I use a lot. What we did was we took generations of designers and started training analysts. The majority of the people today are analysts. They never designed anything in their life. The skills that it takes to do design are, in many respects, very different from the skills it takes to do analysis. We're back to this inverted pyramid. As a designer, you have to make enormously consequential decisions in the absence of perfect knowledge. Learning how to do that is hard. If you're an analyst there are many elements that are basically like academic scientists. It's okay if you study this problem forever. There's no consequence in taking another year to study this problem. So, everything about what you do and how you do it is just different. I would say that we are certainly entering, if not in a period where some of those design skills are going to be required. There have been many epics in the life of nuclear weapons design. The very early period, when we didn't really know very much sort of lasted form '45 up to the '60s. Then, there's the '60s, '70s, and '80s, when much of the modern stockpile was done. And then, the '90s, and 2000s, up until 2015, or so, when we were all about stockpile stewardship, and I think we're entering into a new epic right now — my own personal opinion.

Zierler:

George, can you talk a little bit about your time in Washington in 1989 when you served in an advisory role? How did that come about, and what were some of your contributions while you were in Washington?

Miller:

I moved to Washington shortly after John Nuckolls became director. Interesting side comment, people frequently ask me how I got where I did in my career. I said, "After applying for a job at Livermore, I only applied for one job ever again, and that was to become director when Nuckolls was selected. So, I wasn't selected. Everything else, somebody just asked me to do." So, anyway, let me politely say, there was a certain amount of tension between John and myself when he became director. I think, had I not taken a year off, we probably would have worked it out anyway, and we certainly did. Someone that I knew quite well had been the lead staffer on the joint committee on atomic energy - Mel Grier. He was an advisor to the lab after he retired. He also knew ADM Watkins very well, and Watkins wanted a couple of people who, technically, he could trust. ADM Watkins became the Secretary of Energy basically, in the midst of Rocky Flats problems, and he was going to remake the department to operate much more like the Navy than the Atomic Energy Commission. He wanted a couple of technical advisors, and Mel suggested me. So, ADM Watkins selected me and I went to Washington for a year to advise him. So, in addition to just providing advice to ADM Watkins on a whole variety of technical subjects, my main job ended up being working on a thing called Complex 21, which was the second attempt to deal with the fact that the production complex was falling apart, because much of it was built in World War II. There had been a previous study – the Starbird Study after General Dodd Starbird – in the mid 1970’s, that had done some upgrades. Basically, as people who are students of history know, the Reagan buildup started under President Carter. The budget that Reagan inherited was actually Carter's budget, it started both the nuclear weapons reinvestment as well as other DOD reinvestments. So, the problem of the aging production complex had been studied in the 70’s. This was '89, and it hadn't fixed everything, so there was a very large study tasked by ADM Watkins associated with what we need to do to modernize the production complex. I worked on that study, and basically was in charge of the part that recommended an appropriate size. How big does it have to be? So, there was a team that did that, and I was in charge of that. We made two observations, the first is the size of the production complex in terms of throughput should be the number of weapons in the stockpile divided by the average lifetime, and that production should go on continuously. The second observation is if you have one of every machine that's necessary to make a new weapon, you probably have an inherent capacity in one shift of 100 weapons a year. So, it doesn't take detailed targeting studies to tell you how big production complex could be. You can derive the numbers pretty easily. Now, the details actually do matter, but you can come up with pretty good estimates. So, the sizing criteria for the Complex 21 — and we assumed that there would be a deployed stockpile of about 1500 - 3000 weapons. You can come up with that in any number of different ways, and probably a reserve stockpile of a similar size. So it was likely that the stockpile would be 3000 - 5000 weapons, roughly. The numbers in the START treaty are not terribly different from that. Again, it's not dependent exactly on the details. So, the sizing criteria was one, and then I did another study for ADM Watkins. About that time there was an interest in more intrusive verification for arms control. There was a lot of discussion about whether we could actually negotiate and verify a treaty that had to do with numbers of weapons, so that you could actually verify dismantlement of weapons and numbers of weapons and do that without revealing classified information. I did a lot of stuff on a day-to-day basis, but those are the two big things that I worked on.

Zierler:

In what ways, George, did your time in Washington help prepare you for future leadership at the lab?

Miller:

Two gross ways, or maybe three. First of all, just understanding in detail how Washington works. It works a lot differently than most people think. It was sort of a shock to me to discover that I didn't actually work directly for the Secretary most of the time. I worked for the head of defense programs. He would have a staff meeting at 6 o'clock every morning, and based upon the highest urgent priorities, he'd decide what the staff was going to do that day. So, it was very, very short-term focused, and I had no idea it was really like that. One of the things that I learned very quickly was if they asked for an answer to a question by 3 o'clock in the afternoon, and if you gave it to them at 3:30, they didn't use it, because their timescale had already passed. So, sort of learning how Washington worked was, I think, really important. Just learning the people in a more personal way, many of the people that I met during that year I interacted with for many, many years to come. Again, understanding better the relationship and the tensions that naturally exist because of the civilian/military separation of nuclear weapons.

I mean, that's been around since the Atomic Energy Act, and it's there for good reasons. But understanding the tensions a little bit better, I think, was also really helpful.

Zierler:

When you got back to the lab, in what ways had your perspective changed as a result of your time in Washington? Did you feel like you saw a bigger picture?

Miller:

Yeah, I definitely saw a bigger picture. I understood that the people in Washington are under incredible pressure from Congress, from other parts of the department, from the DOD. I really gained the appreciation of what they were going through. It was incredibly helpful to me, because it changed the way I tried to have the lab interact with DOE Headquarters in particular and the broader Washington national security apparatus in general. As I said, I really understood the incredible difficulty of the jobs that they have — in fact, the broader picture of having those budgets or programmatic decisions across a very broad spectrum of issues, much broader spectrum of issues than I completely understood at the time.

Zierler:

When you were named Associate Director in 1996 for national security programs, post-Cold War, what were some of the major issues with regard to national security that you felt fell under your portfolio that were relevant for the lab.

Miller:

Certainly, in addition to coordinating the various parts of the lab that worked on the nuclear weapons program, I oversaw the work that the lab does for the Department of Defense and the intelligence community. There's always been, as a result of the science and technology that the nuclear weapons program built up, a lot of synergy with some of the problems that the DOD has. At that time, there was a lot of worry about counterterrorism, a lot of worry about nonproliferation. Within the intelligence programs there was a growing recognition of the importance of things like cyber security. So, one of the initiatives that I had was to try to develop a better relationship between the lab and some of the agencies that worry about cyber stuff. Something that I continued when I became a director 10 or 15 years later, was the importance of space, and my belief that the lab had capabilities in space the space arena that could be helpful. A couple of programs that did very well during that period were the lab's biosecurity programs. So, we made a lot of investments in biosecurity. So, in that particular role, because there were still ADs running the programs — I didn't actually run anything. I was more of a coordinator, but because of my longevity and experience, I had more coordinating power than the job description would have suggested. Anyway, one of the major jobs was to decide how would the lab invest its discretionary resources in the national security area, and biosecurity was one. Cyber security was another and trying to advance some of the needs of the work that we did for the Department of Defense. Those were all things that I tried to do. The other job, which was easy for me, was to adjudicate the budgets in terms of how the budget got presented to Washington, and requests for the various parts of the nuclear weapons program.

Zierler:

Now, your work for the National Ignition Facility, was this entirely within the context of Livermore, or was this separate?

Miller:

It basically was my job to make sure NIF got built, as an AD. The hands on project management was done by Ed Moses, who is just an absolutely fantastic person and project manager and leader. So, there were a few things that I had personal perspectives on, but most of the direct work was done by Ed. My job was really, first of all, to rebuild a consensus that you wanted to build NIF. The reason I got the job was the previous AD resigned because of two things. First of all, it was discovered he didn't have a PhD, and second of all, there was a major hiccup — budget overrun and delay in NIF. So, he resigned as a result of principally the budget problems. So, it was not at all obvious that NIF was even going to get built. So, there was a lot of work associated with re-developing the consensus within the three labs that NIF needed to get built. Actually, at the program level, neither Sandia nor Los Alamos wanted it built – at least not the complete project. There were a variety of proposals to only build a subset of the beamlines and limit the kinds of experiments that could be done. These proposals would have completely eliminated any pursuit of fusion ignition. Then there was the huge amount of work to be done in rebuilding the consensus both with the Department of Energy and with respect to Congress. So, most of my work was outside the lab, as I said, rebuilding that consensus.

Zierler:

What were the main objectives of NIF, and how important was it from a national security perspective?

Miller:

NIF is the grandchild of Jim Schlesinger. When the laser was invented in 1960 that's when the national inertial fusion program got started. When Schlesinger as Secretary of Energy founded the ICF program, he said it's there for three reasons. First of all, it's going to give you the ability to do relevant experiments in physical regimes that had to do with nuclear weapons. So, high temperatures and pressures. Second of all, it'll allow you to do incredible basic physics, astrophysics, equation of state, that would be both an attractor in terms of recruiting people, but also contribute to national basic science. And he said that third of all, it has the potential to be an energy program. If we get fusion, you could eventually be able to make energy with it. So, those three legs of the stool have been part of the ICF program since the '60s. So, NIF, in particular, was built based on the recommendation of the National Academy of Sciences in about 1995. Basically, Livermore was pushing to get what was called a high-yield facility, which required 10 megajoules of energy, a 10 megajoule laser. That was the yield at which, given what was known then, that you could pretty much guarantee you could get fusion in the laboratory. The national academy said that's too big a step because the biggest facility at the time was Nova and it was only 40 kilojoules. So, going up that far — actually, NOVA was supposed to be twice the size at 100 kilojoules, but it only half was built, so it was only 40 kilojoules. The NAS judged that the step between 40 kilojoules and 10 megajoules - was too big a step. Their view was that at around 2 megajoules, — NIF is actually 1.8 MJ — there's a 50/50 chance of getting ignition. You may get ignition and will then have hard laboratory data that will allow you to say what is required to get robust fusion burn. So, that's why NIF was built at 1.8 megajoules. The national security implications of NIF, apart from the fusion energy, and the basic science, are that you can do really important weapons physics in the temperature and pressure regimes with very sophisticated diagnostics that will answer questions that are really important about weapons themselves. If you get ignition, that will be really important. I mean, during the period when we were re-baselining the NIF project and the government was deciding whether or not to continue to build NIF, the defense department did a study on their views on the subject. Their comment was the largest single, unknown piece of weapons physics is associated with fusion. So, you should build NIF, hopefully you'll get ignition, so you'll have a platform to help resolve some of these unknown issues that the weapons program has with fusion. So, it's high-temperature, high-energy density physics associated with nuclear weapons, and trying to see if we could get a fusion platform to study weapons. So, those are the two major national security implications. In my mind there’s a third reason; some people place it higher on the priority list. That is that you can create the effects of nuclear weapons, both neutrons and X-rays, and use them to expose pieces of military equipment and understand their vulnerability to nuclear effects. So, weapons effects. So, fusion, weapons physics, and weapons effects are the three national security reasons for NIF.

Zierler:

What has been its long-term successes?

Miller:

There have been some very important results from the weapons physics and weapons effects areas; the weapons physics community has done some really remarkable stuff. There’s also been some really interesting basic science, astrophysical stuff. They actually can simulate some aspects of super novae that the community finds interesting. NIF has also done some very nice work in the weapons effects area. In the whole area of high-energy density physics, atomic physics, hydrodynamics, really great work. The quest for fusion ignition has made some very good progress but has yet to get to ignition – that’s really hard, harder than we expected. I think the current estimate is that they're within a factor of two along a couple of different axes from getting ignition. They've gone up a lot. They have gotten what's called alpha heating. That is, they made enough energy though fusion such that the alphas actually heat the plasma itself from the alphas of the fusion reaction. They haven't gotten runaway ignition, which is still about a factor of two in a couple of important parameters from success. I would consider what they'd done to be really good work. Starting about four years ago, the budget was cut by 30%, so they're making slower progress than they would have expected. They have yet to completely explore the physics of ignition at 1.8 megajoules, is the way I would say it. Whether they get ignition or not remains to be seen, but there are other parts of the phase space that need to be examined before they write the report and say, "We're done." My personal view is that if the budget stays at the level it is, within the next couple of years, maybe five years, they'll be able to say how big a facility you'd need to get robust fusion burn and if a small upgrade is necessary to demonstrate ignition. My personal view right now, is that the facility is approaching 15 years old, and it needs a little bit of refurbishment. It's beginning to wear out, so it could use a little bit of investment, because that's one of the things that got sacrificed when the budget got cut. They haven't been maintaining it quite the way we had originally envisioned. I would say, in the near term, a modest, small refurbishment of things that just aren't behaving completely according to the original specs, and then in something less than five years, there should be a clear indication of what it's going to take, if they haven't already gotten ignition. I still very strongly believe, as a weapons person, that you need a platform to examine fusion.

Zierler:

When you were named Associate Director at Large, did you see this as the prep stage for becoming director, or was it its own entity, its own separate work?

Miller:

The Associate Director at Large position at Livermore has a long and distinguished history. Basically, the Associate Directors at Large, and there have been a few — Mike May, after he stepped down as director, was an Associate Director at Large. Carl Hausmann, an important person in the history and development of the Laboratory was Associate Director at Large. They were Associate Directors at Large for many, many years. That's a position they were in when I came to Livermore in '72. It basically is somebody who's very senior, and advises the director, and does a whole bunch of institutional-like things. The particular plan for the site at Livermore was done by Hausmann when he was an Associate Director at Large. When Mike May was an Associate Director at Large, he established and started the whole institutional Laboratory Directed Research and Development Program (LDRD). So, they typically did institutional-like things. I actually did not think I was going to become director. I thought Mike Anastasio was going to be director for about 5 – 10 years. I'm older than Mike is and that I would probably retire before he did. It was an appropriate senior level position for me. It was time for Ed Moses to become the Associate Director. So, to me, it was more like an end of career, senior advisor position.

Zierler:

Then, eight months later —

Miller:

Yeah, eight months later, DOE decided to compete the contract at Los Alamos. Mike, for reasons I've always admired and never understood, agreed to lead that bid team. I’m not sure he never imagined that they would get selected. At that point, the Board of Governors asked me if I would be willing to interview for director. And I said, "Sure, why not," and they obviously selected me.

Zierler:

Did you think while you were interviewing that you were going to get the job, or was even that a surprise?

Miller:

I guess I would say it wasn't a surprise. I knew everybody who was a candidate. This is not an egotistical statement, but I thought I was better prepared than any of them. I'd been an associate director since 1985. For 20 years I'd served under Batzel, Nuckolls, Tarter, and Anastasio. So, I worked for four different directors. It's hard to imagine anybody who was better prepared for being director than me.

Zierler:

Even with that in mind, in what ways can you truly not be prepared for aspects of the job until you're in it?

Miller:

Absolutely, you cannot. You can't imagine anybody who was better prepared, and I wasn't. I think it really is associated with — as Harry Truman said, "The buck stops here." Mike used to start every morning before anybody came in by standing in his doorway looking out and said, "I know it's my fault. Let's get on with it." So, first of all, there is nobody else to decide. If there is a monumental decision, it's yours to decide. Unless you are really prepared for it, people will try to make you make every decision. So, it's figuring out how to delegate downward and make sure it stays delegated downward. The breadth of things that come up, everything from a personnel decision, major programmatic things, environmental safety and health things, security things, it's just everywhere. Most importantly, it’s your job to think about and prepare the lab for the future – and you have to find enough time for that in the midst of all the short-term urgencies. For me, and I guess Mike had this in spades at Los Alamos, having to craft a bid for the lab, only a few months later was when the contract bid process started, and create a team of people from inside and outside the lab. The co-leader of the bid team with UC was Bechtel. Along with several commercial companies, they had absolutely no clue what the lab was. Creating a team, was nontrivial; there were major decisions about who I would accept on my bid team for senior positions. People I didn't know. I was used to picking people, but they were always people that I had known for ten years. So, just a tremendous breadth of things that I never had to do before.

Zierler:

In terms of your leadership style, did you go in with the specific plan, or mandate, specific things that you knew you wanted to accomplish, or did you want to wait until you were actually in that office to determine the kinds of things that were most pressing for your concentration?

Miller:

I guess I would say there were two answers to that. The first, and this is what I told the Board of Governors when I was interviewed. I said, "It is not obvious to me that a national security, much less, a nuclear weapons laboratory, can do its job of serving the nation while it's a for-profit company. My main job is to do everything I can to try to make that a reality, but I will tell you ahead of time, I am not confident that the answer is yes."

Zierler:

Was this a revolutionary thing to say, in terms of how things had operated for so long?

Miller:

Oh, absolutely. Since their inception, Livermore and Los Alamos had been operated by the University of California. Even though many people thought that the university was a hands off manager, many aspects of the culture of the laboratory came from UC. The thing we talked about earlier, simultaneously allowing dissident views that have significant political consequences to exist within the laboratory and be expressed is part of the UC culture. In the same laboratory, you have people developing nuclear weapons, and you have people trying to figure out how to verify nuclear test ban treaties. Something that was fundamental to the nuclear weapons program, and at the same time, another part of the lab is trying to figure out how to make a treaty verifiable that will ban it. This internal tension was quite natural in the UC system. Whether a company could keep that alive was not obvious to me. I'll give you another personal experience, and this was from an interaction with the Secretary of Energy, Sam Bodman. We were in a meeting and he said, "If you don't do this, I'm going to cut your company's fee." I said, "Mr. Secretary, you do not understand the laboratory. The laboratory is not motivated by fee. You can cut my Goddamn fee to zero, and I'm still going to do what I think is in the country's national best interest. That's who we are." But the fundamental question was, is that ethic going to survive in a for-profit company?

Zierler:

How did you articulate an alternative to this model to the interviewing board?

Miller:

I basically said, "This is the point of view that I will bring as the laboratory director. The ethics and the culture that came from the University of California, I will do everything I know how to make survive. You can still have disciplined operations," which is what these companies were fundamentally becoming part of the partners for. "You can have disciplined operations in the context of this broader ethical point of view, but I'm not going to sacrifice that, and if you're interested in somebody who will weigh those two against one another, pick somebody else."

Zierler:

Obviously, they liked what you had to say.

Miller:

Yeah, they certainly picked me. I think another part of it that's important in the structure of the Board of Governors is that the chair of the board always is a University of California regent, and even though the fee gets divided up in all kinds of funny ways, the chairman always has 51% of the votes. So, if it comes down to a fundamental question of the variety that I was discussing, I had confidence that I would have the backing of the University of California board members. In particular, the number two person was Senior Vice President at the University of California, Bruce Darling, who went on to lead the National Science Foundation, I think. Bruce is just a wonderful, wonderful person. He was always somebody I could talk to, and he was always somebody who I knew had my back where these kinds of things were concerned. I was hopeful that things would not go south in particular situations. But they haven't, so that's good.

Zierler:

What do you see as your major achievements in your tenure as director?

Miller:

I guess probably the most important one was we got through the contract transition.

Zierler:

You survived.

Miller:

Within six months of the contract transition, we had the first layoff. This was 2007, or 2008. We had the first layoff we'd had since 1973. We had to because of the combination of the budget cuts that were occurring in Congress, and the additional fees that we had to pay to the new company. I don't think anybody would describe that as an accomplishment, but having the lab survive that, to me, was an accomplishment. I mean, it was just incredibly painful. It was personally painful. I went through name by name, the list of everybody that was being laid off multiple times, and every time I'd ask questions about why the particular managers are doing this. Only in a small number of cases did I overrule them, but it was very painful. Then, just the whole contract transition — transitioning salaries, positions, everything was really difficult. Again, the lab came through swimmingly. I think another thing we managed while I was director was to get NIF finished. That's really the whole lab pulling together and mostly due to Ed’s superb qualities as a manager and leader, but the fact that we were able to keep the support up, I certainly contributed to. And one of the other things that I've always felt is uniquely the lab director's responsibility is to think about the future. Again, depending upon where you are in the organization, you may be 40/60, or 30/70, or 90/10 worrying about the urgent, near term and your individual organization. You know, what you're doing today and tomorrow, but somebody's got to worry about what the future is. I'm pleased with the things that we invested money in to try to build up parts of the lab and again, generally speaking, I think we made good investments.

Zierler:

To what extent did you concern yourself with broadening Livermore's involvement with civilian initiatives?

Miller:

I tried really hard to push that. That was one of the things that I invested in. The open campus, having a part of the lab where it’s industrial level security, not government, classified level security. Having that was something that I initiated. I supported other efforts and interacted with the California Public Utilities Commission to try to establish a long-term partnership between the lab and the utilities in California through the PUC. We proposed many areas of cooperation including better modeling of the grid and cyber security. It's a very difficult problem to model all of the interruptible green sources of energy with all of the sustained base load. That's not a simple modeling problem. It’s also important to better prepare for the external hacker environment that's attacking our infrastructure all the time. The lab has expertise in many areas that are relevant to issues facing the utility companies. So, pushing for long-term relationships there was something that I established and both of them are going along very well.

Zierler:

In what ways did institutional or bureaucratic inertia prevent you from accomplishing all that you set out to do as director?

Miller:

It was a fight all the time. I had to spend time fighting bureaucratic wars that I would have rather spent doing other things. It's just incredibly difficult to get things done. One of the things that turned out to be just an incredible mess was, because of my time in NIF, I wanted to invest in advanced laser systems. Laser technology for the DOD, for the civilian sector, for basic science, but also for preparing for the time when we would get fusion in an ignition sense, and we would want to start thinking about how to turn this into a power plant. Every time I turned around, some bureaucratic problem would erupt, in terms of how I wanted to invest the money. Most of it from Washington or Congress, and some of it from the local site office. Some from within the laboratory who thought I was putting too much emphasis on NIF. They saw it as augmenting NIF as opposed to investing in the future.

Zierler:

George, I know you mentioned before, "The buck stops here." Who were the people that you relied on either formally or informally for advising? Would you go to fellow lab directors elsewhere? Did you have a kitchen cabinet? Did you rely mostly through the official channels of your ADs? Who were the most important players in terms of people that you looked at for their input and insight?

Miller:

So, in terms of laboratory directors, I talked to Mike Anastasio all the time. He and I have had a variety of relationship. He worked for me, I worked for him, we had complimentary positions. I've known Mike a very long time. He's a very good friend, and I really value his views and input. He's somebody I talk to a lot. Bruce Darling from the university is also somebody who's very insightful really about everything. He's insightful about state politics, university politics, personnel issues, all over the map I would talk to Bruce quite a bit. I guess I would say I had an unofficial kitchen cabinet. Some of them were in official positions. The second deputy director I had, Tom Gioconda is wonderful on a broad range of topics. He was a retired Air Force brigadier general that I had gotten to know because he was in charge of NNSA when we were trying to re-baseline NIF. I hired him to become Deputy Director. Tom was just a wonderful person with knowledge about all kinds of stuff. So, I would talk to Tom a lot. Then, my other Deputy Director, and in particular, I talked to other associate directors. Ed Moses was a key advisor on almost everything – very thoughtful and insightful. I talked to Bruce Warner who had been a deputy in NIF, a deputy in Global Security and AD At-Large. After I retired, he became AD for Global Security. I consulted with Jan Tulk who had been a lab lawyer and in HR and spent a lot of time with my CFO Linda Rakow. A very close associate was Larry Ferderber, my chief-of-staff. He had worked in a variety of places in the lab and also as a staff member to Senator Harry Ried. So, just a variety of people. I tended not to just work in the official structure. Any time I wanted to accomplish something; I've always worked through the official structure. But generally speaking, it was just people I know and who I had come to rely one. The one Bechtel person in addition to Gioconda that I relied on a lot was the person who was the head of operations and business - Frank Russo. I found him to be very knowledgeable, who figured out the lab quickly and fit in while bringing a new perspective - just an all-around good guy. If he had to do something because Bechtel told him, he would let me know, and he would always give me good advice and then say, "I can't do this because of Bechtel's whatever." I appreciate people who are like that. A thing I learned very early in my career, and I tried to always do is I try to have people around me and listen to people who are different than I am. I frequently tell people I usually, not always, know what I think about something. But I want someone who challenge me and will think about a problem differently. So, that was always something that I tried really hard to do and value. People who are just "yes" people, I don't care for.

Zierler:

I was going to ask you, by the time you became director, you were of course very far off from your early days in the design lab as a physicist, but I am curious if your training in physics, and the way you thought about the world though a physics lens informed your style even as director.

Miller:

Yeah, I would say so. I valued the experience and training in physics and how you think about a problem, and how you try to break it down and dig down into the details, and understand the details as a way of understanding the broader issues. As I said earlier, the organizational skill I learned earlier in doing experimental physics, all of that was really important. The other, which had nothing to do with physics was going to William and Mary. It's a small liberal arts college, and they made me take history, and government, and philosophy, and economics, and all that soft science stuff. I read as much of that stuff today as I do technical stuff, and so having that background was just incredibly valuable to me, particularly as I moved up in the organization. I would have been a lesser person had I gone to a purely technical school.

Zierler:

Sure. George, when you knew it was time to step down, talking about your responsibilities as director, thinking toward the future, what were the aspects of your accomplishments that you felt were durable and that would stay in place well beyond your time there, and what were the areas where you were concerned that despite your best efforts, problems would crop up in the future that you were aware of but there wasn't anything that you could do about it?

Miller:

I think in the latter category, there's just a whole series of what I would call bureaucratic incursions. The system just kept becoming more viscus. As part of this aspect, which really had to do with part of why I became director, is that I think of the DOE laboratories at large, particularly Livermore, Los Alamos, and Sandia, as partners with the federal government. We're federally funded research and development centers. FFRDCs. And FFRDCs have a partnership role with the federal government. Increasingly, DOE, during my tenure and afterwards, began to treat the laboratories as contractors. That was my Secretary Bodman comment. "Just do what I say, or I'm going to cut your fee." The current head of NNSA, Lisa Gordon-Haegerty, I think, is trying to change that. But bureaucracy is like entropy. It only goes one way. It's really hard to turn around. So, of all of the things that I tried to fix, I guess I would say I was unsuccessful is that one, and I think it's getting worse. I think the sense of support for national security as a special responsibility, that our job is to give the government the best technical advice we can independent of the consequences is still strong withing the Laboratory. I think that's still well-established in the laboratory. So, that part of getting through the transition we got through, and I think is still very much there. I'm very, very pleased with that. There are lesser things that I think are good. I think some of the areas that I invested in as director, invested in space technology, cyber security, and a variety of other areas have worked out well. The laboratory has maintained its position in some areas I had tried to expand but for instance it's only maintained its position in engaging in climate research. I wanted it to be larger than it was. It really isn't, but it's still a very important position which I'm very pleased with. One initiative that I had that I wished had worked out, and it hasn't, is I thought that in addition to modeling the atmosphere, that we needed to spend more effort modeling the Earth. So, geophysics has not taken off. But, again, I still believe that, in many respects, the Earth is more complicated than the atmosphere. The atmosphere and the Earth and the oceans are coupled, and if we're ever really going to make more detailed sense of this, we need to figure out all of them and figure out how to couple them better. So, I mean, there's details like that, but again, I think that the one big thing is we got through the contract needle with the UC culture still largely intact. That was probably the most important thing to me.

Zierler:

Well, George, we talked about your work in recent years at the beginning of our talk, so we're current up to the present in terms of your activities. So, for my last question, I want to ask — a theme of our talk today has been that it's important for national laboratory to be nimble, and to be able to evolve overtime as broader situations change. So, obviously, we can't predict the future, but using your powers of extrapolation, and your uniquely singular institutional history, essentially having spent your entire career at Livermore, what do you see as the most productive and positive way that Livermore can continue providing scientific and national security support and services to the United States going forward?

Miller:

I think it is, to be broad based, but focus on some key capabilities, and then apply those capabilities wherever the problems may be. Presidents change at least every eight years, if not every four years, and their perspectives change. Whether it's Reagan and SDI, or the issue of Clinton wanting a test ban, national policies do change. I hope we're going to have a president who really turns the technical and scientific apparatus on the problem of climate change. I think the lab can be positioned to respond to these. We have responded to the current virus pandemic. When 9/11 happened, we were ready with technology and people to go work on the problem of terrorism. A lot of the national security environment is changing rapidly, so it's first of all, nimbleness, as you said earlier. That's the key. I think maintaining the kind of organization that allows us to turn on a dime organizationally is really important. I think that's an organizational strength that we need to maintain, and I think we have. Having key capabilities that we can turn to, again, the modeling effort, diagnostics and sensors, but also a lot of the basic science. I think those are the keys, because you can't predict the future. That's one of the regular pieces of advice I try to give the Defense Department. You can't predict the future. All you can do is be agile and nimble like commercial companies are, and when something happens, be ready to respond to it. I also believe that the Lab continues to a have a special responsibility – that has to do with nuclear weapons, their importance in maintaining the country’s strategic deterrent and understanding how to combat terrorism and proliferation. As part of that responsibility, the Lab Directors continue to have the job of annually assessing the health of the country’s nuclear stockpile and reporting their assessment to the President. That’s an awesome responsibility

Zierler:

George, it's been a pleasure speaking with you today, and it's such an incredible value to get your perspective because it's really hard to understand how these national labs work. You've provided so much valuable information for understanding all of the perspective, and your commitment to the lab, and your commitment to American national security. It's really admirable. It's an honor to have spent this time with you, and I really want to thank you.

Miller:

Thank you. I've enjoyed it. It's been very easy conversation. I hope it's been helpful.

Zierler:

Thank you.