TAI: Good morning, Andy, and thanks for agreeing to this exit interview for The American Interest.

Andrew Marshall: Happy to do it, Adam.

TAI: I’ve just had a look at The Last Warrior, a book about you by Andrew Krepinevich and Barry Watts. The authors claim that you’re a humble sort and don’t like to call attention to yourself, and I know from my own experiences that this is true. Therefore, I’ll not ask you any personal questions for the record, and I don’t want to ask many questions about the past; it’s all there in the book. I’d prefer to talk with you about what you’ve been doing for the past forty or so years, which is to try to peer into the future.

But first, please, a meta-methodological question. You have training in statistics, practical engineering, and economics. Yet you’ve made a point throughout your career of being skeptical of claims from authority, skeptical of claims to precision, and especially of claims based on rational-actor models where they really don’t apply. So the net assessment method you developed, and for which you are justly renowned, has one foot in the world of the harder sciences and mathematics (in the form of statistics), and yet insists on a kind of artistic balance between the method and the way it’s to be applied. Can you talk a bit about this balancing?

AM: Net assessment was designed to achieve exactly the balance and inclusiveness you describe. One of the reasons for the switch from trying to do it at the NSC to Defense was that it was designed to help decision-makers think about the decisions they had to make about future U.S. force posture. There was a real customer for it at DoD and, particularly in James Schlesinger, an appetite for this. The NSC lacked the time and staff to do this right, so over to DoD the task went.

At the time, we were trying to look five, maybe ten years ahead, and the focus was decisions about the future force posture, not about political crises or other developments shaping the direction in which global dynamics might go. It was designed to give top DoD leadership a more detailed, rich comparison of ourselves and the Soviet Union—allies too, but the center of it was ourselves and the Soviets.

To do that required several steps. It could not be done in one full, single, comprehensive look. You had to break it down into some geographic areas, or some warfare areas, and then within that to look in great detail at the numbers of ships and vehicles and so on—but just as important, also at differences in doctrine, operational practices, training, and other less tangible or easily measured factors.

From the beginning I believed that to practice net assessment, we needed to look back twenty years to make reasonable projections ahead five to ten years. You had to understand in great detail how we got to where we are. Some underlying change was taking place, things were steadily happening, and only by looking back could you get some insight into patterns of change and so on. So there were several kinds of balances to strike in order to, at the end, have a many-faceted assessment of how we stood relative to the Soviet Union.

TAI: I know you’ve always been interested in and respectful of history. Of course, anyone who’s studied history enough knows that the rational-actor model doesn’t work predicting the past, let alone the future.

AM: That’s one of the reasons I left economics, once I discovered that there existed a field called mathematical statistics… If I had thought at the time, way back when, coming out of the Depression, that a person could make a living as a mathematician, I probably would have tried to do that. But I became disenchanted with economics, even at the University of Chicago where I was. Those were the days when economics tried to develop high theory, and with it came pretensions to be all-knowing. The work of people like Herbert A. Simon and a few others were seen as a minor heresy within economics.

TAI: In a way, perhaps your timing was bad. Now we have behavioral economics, the re-ascendance of microeconomics in a new form which seems to me a very good thing. Schlesinger was an economist, of course, but I think he understood that theory was only so useful, and that microeconomics had to be a part of the mix—and that’s the part where rational-actor assumptions break down fastest.

AM: I don’t know why Schlesinger was asked to come to RAND for the summer of 1962, but very likely it was because he had written a book in 1960, The Political Economy of National Security, one chapter of which picked up on the work Warren Nutter had done leading to some skepticism about the purported growth rates of the Soviet Union.

TAI: I do want to get out of the past, but since we’re talking about Jim Schlesinger—and since you met and knew practically everybody, from Bernard Brodie to Albert Wohlstetter to Herman Kahn, just to name three—I’d like to ask about a less well-known figure from the RAND days, but one I’ve always considered a great creative genius: Nathan Leites. In a Skinnerian era, when Enlightenment universalism got matched up with positivist social science and went to absurd extremes, Leites understood that rationality has to sit within a reservoir of culture and social structure, and that it differs from place to place. So what was he like?

AM: He was wonderful. He did a number of projects for me after the Office of Net Assessment was established. Schlesinger really liked him, too, and had him do some things, even when he was the Secretary. Schlesinger sent Nathan to interview people, far and wide, to get an impression of what their real thoughts were. Aside from being a shrewd psychologist, Leites had tremendous language skills; he knew six or seven languages fluently, both oral and written.

Most people know about his operational code work, but less well known is a book he did on movies, which came out around 1950, co-authored with Martha Wolfenstein, a psychiatrist who was then his wife. It’s an analysis of all the movies that played in New York, I think, in 1947. What’s interesting is the differences that come out between the British, French, and the American movies, with their typical plots and characters like the “good” bad girl in American movies, you know, the woman who seems bad, dangerous, but underneath is really good. But more than that, the book is full of general insights—a very creative effort.

TAI: I never heard of it. That moves to the top of my list now.

Let’s finally get back to the future, to overuse a phrase. You were one of the first people to take the Chinese seriously as a potential adversary or competitor of the United States, and that was back in the mid-1980s, before the Soviet Union dissolved.

AM: We tried to persuade Fred Iklé [Undersecretary for Policy, 1981–88] in 1987–88 that the Russians were really finished and that the next thing we might have to worry about was China.

TAI: Right now, though, I’m as concerned about China imploding as I am about its continuing to wax strong and become a threat. When you look at China’s trajectory over the past 15 years, allowing for the dangers of extrapolation, what do you see happening over the next 15?

AM: I’m of the view that in economic terms China is not going to be able to continue to grow at anything like the recent pace. That’s already happening. A key person I’ve read on this is Michael Pettis [professor of finance at Guanghua School of Management at Peking University in Beijing]. He has been arguing that they’re going to have to give up on the particular model they’ve had for growth. At one of their recent major meetings they indicated a desire to increase the proportion of their GNP devoted to consumption. Pettis says the logical implication of that is that the growth rate has to drop to between 3 and 4 percent, down from the 8 or 9 percent it has been for some time.

He has also pointed out some statistical issues. If the Chinese build a plant whose real worth may be a quarter of what they invested in it, they put it in the books as worth the full value of the investment. Pettis would start by knocking something like 20 percent off their current estimate of the size of their GNP, and then predict growth much more slowly from that lower base.

I’ve been interested more in the implication that their military budget will also not be able to go up by 10 or 12 percent per year in the future, as it has been doing in the recent past. Also, China, in contrast to the Soviet Union, is not autarkic. It will have to be importing energy, food, and other essentials for a long, long time. So they have a fundamentally different set of strategic problems than the Soviet Union did. ONA was never alarmist about China. We just thought it was a good idea to spread around our attention some, and that China was the most likely candidate as a major challenger.

TAI: Personally, I’ve never lost any sleep about the Chinese. What worries me is how American political leaders react to stimuli of various kinds from overseas. We seem to have a knack for underreacting and then overreacting and then underreacting again. Not that there’s anything we can do about it, seems to me.

AM: That’s why it is so important to have some kind of reliable empirical base for thinking and planning, so that at least some people in the government can keep a foot in reality. My idea at the beginning about China, given the predictions that [longtime senior RAND analyst] Charlie Wolf had made about future GNP and likely growth, was that China would grow substantially, and that it was just normal that it was going to put some of its increased output into the military, and, starting from a pretty low base, it would not be tightly bound to legacy investments. This assumption was reinforced by a study we had done probably around 1994, not long after Andy Krepinevich had done his important study on the potential revolution in military affairs. We knew the Russians were thinking and writing about the implications of new technology on military matters, so we wondered who else might be doing that. We started a project to read the military journals of a number of other countries, and we found that, yes, at least one other country was on to the theme and writing a great deal: China. Moreover, they’re unique in expressing an ambition, by the middle of this century, to match or be better than the Americans.

TAI: There’s an echo chamber out there, for sure. For example, when the Office of Net Assessment was set up, and people realized what it was doing and began writing about it, the Russians, with their famous correlation of forces concept, were surely paying attention. I don’t know if they changed their operations and methodologies of assessing us, but I wouldn’t be surprised. There’s an echo chamber with the Chinese, too.

AM: Yes, I’ve had good press in the Soviet Union and in China, though I never visited either one.

TAI: Now, ONA actually did real planning. We understand why the military needs to plan, because the lag times for weapons systems are far beyond one presidential or senatorial term. But in the rest of the government, there doesn’t seem to be much of a culture of planning anymore. I don’t know if there really ever was one, but I know that when George Kennan and Paul Nitze ran Policy Planning in the State Department, they actually tried to plan. When I was there, it was the last thing on anybody’s mind. Aaron Friedberg has told me stories about trying to set up a planning function in the Vice President’s office, a frustrating experience as he relates the tale. Did we use to plan better as a government? And if we did, why don’t we anymore?

AM: It may be more strategy than planning that has been lacking more recently. I’ve had Scooter Libby looking at this. Strategy—not just planning, exclusively—seems to have disappeared. I think we once did better, certainly under Eisenhower. He was impressive, I think.

TAI: Well, there was Solarium, for one thing.

AM: Solarium, yes, but also, from time to time, groups of people would be assembled to look at some issue. The Gaither Committee was one, but there was a series of those leading up to it, some with different focus.

I think some of the change has to do with generational experience. People who are 15–20 years older than I am, who lived through the 1930s and then World War II and saw the success of the planning then, also looked back to see that with World War I things had gotten screwed up because we hadn’t planned or foreseen enough. So at the end of World War II there was an effort, including the efforts to put into place the United Nations and a lot of other things, not to make that mistake again, to be more focused on looking ahead. These people were some of the main actors in the government for the following twenty years or so, well into the late 1960s. It kind of evaporated after that. We no longer seem to have the generational awareness that kicking things down the road often doesn’t turn out very well.

TAI: I picked up E.H. Carr’s The Twenty Years’ Crisis a few days ago—I don’t know what made me do it—and he has this famous statement about the utopianesque disaster of the 1920s, the Kellogg-Briand Pact business, the spirit of Locarno, and how all that led to the very ugly 1930s and then to war. I couldn’t help thinking that the sort of “end of history” triumphalism of the Clinton years—the “perfect storm” of American power and values coming together—was another example of a mindset that dispensed with thinking about strategy because it deemed it no longer necessary.

AM: Right. It was a way of feeling that went well beyond the so-called peace dividend; it was a feeling that we could just relax because the big stuff was set. That’s not how serious policy people in the government thought, of course. I’m talking more about the political class and the intellectual elite.

TAI: To me there’s a parallel between the way Carr described the 1920s and the post-Cold War era, at least until about 2007–08, if not already September 2001. Things are getting a bit edgy lately. I’m thirty years your junior, Andy, but I see more powder kegs around the world, some of which are already lit—more serious trouble spots out there than at any point in my lifetime. You could easily make a list, too, and we both know these dangers are not one-offs; they compound one another, they cascade into an avalanche of uncertainty. And people sometimes do stupid things when they’re panicked and uncertain. Are you worried? Some people say everything’s fine right now, that the chance of hegemonic war is lower today than at any time in modern history—and there’s certainly some evidence for that claim. How do you assess the “fright scale” now, where some people say the world’s going to hell in a hand basket and others say things are a little cluttered but really not so bad?

AM: I agree with you that there are many troubling things going on. But, in contrast to the Cold War, there’s no immediate existential threat. There are lots of bad things going on, yes, but I guess I fall into the let’s-not-panic school.

TAI: Here’s a related question. Maybe a dozen years ago or so, it became popular in academe to announce the decay of state power because of the increasingly poor fit between the traditional Westphalian unit, the state, and the way that economics and culture were developing both below and beyond the level of the state. Some contended that this meant that the quality of the state was actually more important than ever, because it needed to manage a more complex environment. Others lauded the development because they were anti-nationalist at heart. I was a skeptic at first, but now I’m a believer. I don’t think one can deny the porosity postulate anymore. The weakest states, the ones most artificial because they were superimposed by colonial powers on non-Western societies, barely met the Weberian mean, if they met it at all—and these are crumbling. Look at the Middle East, where four states have pretty much stopped existing. In your career, for more than forty years you could basically take for granted the state-centric nature of the system. What do you think now?

AM: We had some interesting work done on that very subject, the artificiality you point to, and the likelihood that a number of these states may disappear or break down. Some people have been thinking this way for a long time. I remember a Frenchman talking to me maybe thirty years ago, mainly about sub-Saharan Africa, about the artificiality of the existing state system. He said there would have to be a second wave of de-colonialization that would rearrange borders to accord better with tribal and other more naturally cohesive groups as the new states.

TAI: What’s happening in the Levant, and in the Middle East and North Africa generally, has led some to posit the crumbling of the Sykes-Picot system. But that’s a mistake. They’re looking at a symptom, not a cause. The cause is the decay of the state itself; the state system’s dysfunction is epiphenomenal. You see it with Mali and the Tuareg rising after the Libya adventure of March 2011. I take this pretty seriously partly because we as Westerners don’t have a lot of intellectual practice thinking about a world that’s not state-centric. Some have called this the new medievalism, because before the modern state existed there were all kinds of overlapping institutions wielding social authority. We’re not practiced in thinking this way.

AM: I am not sure about the future, but clearly the Westphalian system is under pressure. It brings to mind Chris Ford’s book, The Mind of Empire, which suggests that the Chinese may be another group that finds the Westphalian view alien.

TAI: Sounds like we need more work done on China’s operational code, then.

AM: We might. Future generations might not be happy if we don’t do that work.

TAI: Speaking of future generations, consider that if you look at old issues of Life and Look magazines, you’ll see famous actors advertising, say, the health benefits of cigarettes. Recently I saw a documentary on how long it took us to figure out what tuberculosis was and how to cure it. It reminded me that after Koch discovered the germ theory of disease the majority of doctors refused to accept the evidence. Lots more examples are easy to cite, so we regularly look back on prior generations and wonder how they could be such idiots. So here’s the question: When future generations look back twenty or thirty years from now onto our time, what’s one or two things they’ll say we were a bunch of dopes not to realize?

AM: Some of the work we tried to get done more recently on future security environments was precisely on trying to pick up areas where the current beliefs were likely wrong. One that we started with around 15 years ago was this notion of peak oil, which was wrong. There’s a lot of methane and other hydrocarbons in the mantel of the earth. Or take cropland. There’s a general sense out there that urban sprawl is destroying a lot of needed cropland. But in the United States and some other places, the greatest extent of plantings was maybe forty years ago. The productivity of the seeds and so on has increased at such a rate that we’re using less cropland to grow more food than ever. And the same goes for copper supplies and lots of other resources people have worried about. Studies are underway on these common assumptions to find out which ones make sense and which ones don’t.

TAI: Well, we do have this “end of the world” syndrome that comes along every once in a while in human culture. We had the Club of Rome “limits to growth” meme back toward the start of the ecology era, and we revisit Malthusian doomsaying on a pretty regular basis. It’s just the way we are. Whenever we get nervous about things in general, we go into “end of the world” mode.

People are pretty nervous these days, and the reason is, I think, that the capacity to believe innocently in any cultural or religious template has become more difficult. My TAI colleague Peter Berger has written much about pluralization, by which he means simply the process of people realizing that others think differently about things they hold dear. As soon as you realize that, you turn the light inward, and ask, “Well, why do I think what I think?” The self-reflection that happens next can be unsettling. This is happening in the world now in a much deeper and broader way than ever, and it creates a lot of ambient anxiety and hence “end of the world” thinking. That sound right to you?

AM: It does. If populations are less passive and more restive, that affects how others control their own populations, and so it should figure into how we would go about deterring or fighting if we have to. Obviously, if we’re going to intervene in a country we need to know a lot more about exactly how those people think. We need to do the anthropology and keep it up to date.

TAI: As you know, the U.S. military education system tries to do that. Anna Simons and her colleagues at Naval Postgraduate School perhaps succeed to some degree, but it seems to be an uphill battle. A lot of military guys think by default that since we have the bigger guns and the firepower and massive technological advantages, we don’t need to know how others think.

AM: It’s one of our major failures, that we don’t study other people, other groups, more than we do.

TAI: Let’s move from failure to frustration. The Europeans are our best friends, but they can be awfully frustrating sometimes. The European Union never seems to consolidate itself so that it can stop looking inward and be a more active partner with us around the world. When you look at the future of Europe, do you see a kind of perpetual limbo? Do you see it collapsing and regressing, or moving ahead? Today we have the Greek crisis, Ukraine, jihadism within France and other countries, just trouble everywhere you look. It’s easy to be pessimistic, but is that right?

AM: I haven’t thought a lot about Europe lately, but the basic question you pose is nothing new. In the mid-1950s people in RAND’s social sciences division would go to Europe and come back with various tales of why the Europeans could not live up to agreements they had made about the number of forces they were going to put up, and even to claim that they couldn’t afford it. I used to point out that they’d already gotten to and well beyond their prewar GDP per capita levels—so what do they mean when they say can’t afford it? They could afford larger armies with smaller economies before World War II, so this was obviously not about objective capacities. Something of a different sort had happened to them, to their heads, to their social cohesion, and so on—and I don’t see much of a change since then. So whatever their crisis of the day may be, I don’t think they’re ever going to be a lot of help to us, frankly. Even if the Europeans do achieve some kind of greater union, it’s not clear that it’s going to be a substantial actor in the rest of the world. Economically, they’ll still be important, but they’re still going to be punching under their weight strategically. Except, perhaps, the French.

TAI: So it’s to Asia we should properly look?

AM: The idea that we have to focus more on Asia is right, given that when you look at the whole of the global economic product, Asia, broadly, is increasing proportionately to the rest. In some ways, then, we’ve got to focus on Asia and look for new allies.

TAI: And it’s a different system; it’s much under-institutionalized compared to Europe.

AM: It’s a huge area, and that’s maybe a reason that they don’t have traditions of alliances and multilateral arrangements. In the past, at least, what help could say, Indonesia, be to other people thousands of miles away in, say, Mongolia?

TAI: One of the most interesting phenomena to look at, I think, is how those old constraints of distance are evaporating, at least some. It’s interesting to speculate about the extent to which India will look to its west and become a major actor in the Persian Gulf and beyond.

AM: I think so, too, but the distances still matter. There are so many Indian workers in that area already. I remember talking with an Indian about what action they would take if large numbers of Indians were threatened in some way. It’s clear that the Indian government would not do nothing.

TAI: One more question about the future. My whole professional life has been in foreign policy and national security matters. I was happy with how the Cold War wrapped up. I’d say we won, with not a shot fired; pretty good. All that time I assumed other people were keeping the domestic ship headed in the right direction, only to find in recent years that it isn’t so. I’ve come to believe that if there are existential threats to the United States looking ahead, they’re mostly internal. You surely know as well as I do that that would be in keeping with the pattern of great empires, which have rarely fallen simply because of external power brought to bear. They fell because they rotted from within for one reason or another. I wonder if you share that view.

AM: I do, and I’m concerned about that, yes.

TAI: What do you think is at the core of the matter?

AM: In some deep way, I don’t know, but the education system is obviously a big problem. And then the development that Charles Murray talks about in his book Coming Apart, the growth of an underclass and the undermining of traditional values.¹ There are programmatic solutions for some of these problems, but some of it is deeper, attitudinal, and it’s hard to apply public policy to that. I don’t know how a government gets at that.

TAI: Me neither.

It was 42 years that you were at the Pentagon, right?

AM: Yes, about that.

TAI: You can get used to a place in forty years, the A-Ring and the office and all. Leaving is probably a little disorienting at first. So what’s your next career move, Andy?

AM: One of the reasons I retired, Adam, is that I’m 93. I was beginning to feel a little run down physically. So I don’t plan a second career.

TAI: What? You’re not going to write a book?

AM: That’s too much work. I am planning to write a couple of papers.

TAI: Not classified this time, I hope. Well, I look forward to them because I’ve got a magazine if you need a place to publish them.

AM: I have to write them first.

TAI: Sounds reasonable, I guess. Thanks for the conversation. I enjoyed it as usual.

AM: Me, too—and you’re very welcome.