Search videos...

Daniel Roos ’60, SM ’63, PhD ’66

Search transcript...

INTERVIEWER: Today is June 29, 2010. I am Karen Arenson. We are speaking with Daniel Roos, the Japan steel industry professor of engineering systems and civil and environmental engineering, the founding director of MIT's engineering systems division, the former director of MIT's Center for Transportation Studies and its Center for Technology Policy and Industrial Development, and current director of the MIT- Portugal Program.

He earned three MIT degrees, his Bachelor's, Master's, and doctorate, all in civil engineering, in the 1960s. He is also a co-author of The Machine That Changed the World, a now classic study of the automobile industry that has been published in 11 languages and sold over 600,000 copies. He also serves as chair of the Engineering Systems University Council, a relatively young organization of universities with engineering systems programs of many flavors.

Dan, thanks for talking with us. You started your career by earning these degrees in civil engineering, and you're still officially part of that department, but you seem to have taken on a variety of projects that are more about public policy than civil engineering. Do you still think to yourself as a civil engineer?

ROOS: Not primarily. I followed a career path that many MIT faculty follow, which is about every seven years, I changed my direction in terms of what I was doing at MIT. I've always had an interest in transportation, which is part of civil engineering, but as you suggest, that interest has been more in public policy. What has been a common thread throughout my career at MIT has been an interest in societal problems which require an interdisciplinary approach, bringing together engineers, people from management, people from the social sciences, which is the reason that I and a number of other faculty created the Engineering Systems Division, which right now is where I have my primary appointment.

INTERVIEWER: How much has civil engineering changed since you entered it?

ROOS: It's changed in a phenomenal way. When I first entered civil engineering, the curriculum consisted of engineering drawing, surveying -- they had just eliminated the summer surveying camp, which was up in Maine, but we still took a course in surveying -- and then very traditional courses in strength of materials, concrete, and being interested in highways, we took courses in bridges and pavement design.

And now if you were to look at the curriculum in civil engineering, it's no longer civil engineering, it's now civil and environmental engineering. And I would say at least half of the department, if not more, is really focusing on the environmental side.

A student interested in transportation would take none of those courses. As a matter of fact, none of those courses still exist at MIT. They would take courses in transportation planning, in which they would be required not only to look at the design of transportation facilities, but also what the demand for transportation will be, which requires an econometric approach to build demand models. They would look not only at highway design. They would consider public transportation, and they would consider the social impacts, the environmental impact, the energy impact, of the transportation system. So really just a totally different approach.

INTERVIEWER: What about civil engineering grabbed you initially? In other words, did you like the idea of surveying and drawing and so forth, or was it something else?

ROOS: My father was a civil engineer. And, as I say, I was always interested in transportation, so that was the department that offered the most opportunities in transportation. And I very much enjoyed my education in civil engineering. You know, MIT has always been a reasonably flexible place, so that even though I was in civil engineering, where I was taking my doctorate, I took many, many courses in Urban Studies. And I took many courses in the Sloan School of Management. So it really was an interdepartmental degree that I have, even though my diploma says civil engineering.

INTERVIEWER: Is that to say that the people who were supervising your doctorate were flexible people or that they understood and were interested in some of the same issues you were, and understood the need for breadth?

ROOS: Well, the answer is yes, but I should expand upon that, because I found myself in a very interesting situation, in that I had no intention of pursuing a doctorate degree. I was in a master's program, and at that point in time, not only interested in transportation, but I was very much interested in computers. Computers were really at the point where they were emerging at MIT. I had taken a computer course as an undergraduate and I recognized that I really needed to focus more at the graduate level.

So for my Master's degree thesis advisor I had a very young professor. He had no PhD, he was untenured, and he was just an associate professor. It so happened at that point -- I didn't realize this -- but the Department of Civil Engineering was looking for a new department head. And I gathered that they had offered the job to three people outside MIT, all of whom turned it down, and almost in desperation they went to this 30-year-old untenured associate professor -- would never happen today -- and asked him to be department head, and he accepted.

INTERVIEWER: And this was who?

ROOS: Charlie Miller, who it turned out had a very major impact on MIT. It was a very wise decision. He really revitalized civil engineering. As a matter of fact, I was told that when Gordon Brown offered him the job, Gordon Brown basically said to him, you will turn this department around in five years or we will eliminate it. So, there was a real question about the future of civil engineering.

Charlie came to me and said, I have big plans for this department and those plans to a large extent focus on bringing the computer in as a problem-solving aid. I would like you to stay on to do your doctorate here, and not only that, I have this big, new project in mind, called ICES, the Integrated Civil Engineering System, and I would like you to supervise that.

So, I found myself really sort of shelving my interest in transportation and focusing full-time on the ICES project. Not only was I not planning to do a PhD, I was not planning to join the faculty. Having got involved in ICES and really gotten very, very excited by what Miller was doing, when offered the opportunity to stay as an assistant professor, I accepted.

INTERVIEWER: It's interesting that somebody who didn't have a doctorate himself was encouraging you, then, to get your PhD. Did that seem like a question in your mind?

ROOS: No. I think first of all, he wanted me to stay at MIT, and that was really the only logical way to stay at MIT. And MIT was going through a transition, in the sense that he was not the only faculty member in the Department of Civil Engineering without a doctorate. Because, to a large extent, up until that point, engineering was really practice- and design-oriented. You know, we were just entering the era to a large extent brought about by Gordon Brown where the focus shifted very much to engineering science. From then on, to be a faculty member at MIT, you had to have a doctorate. So, it really made sense. He recognized that.

INTERVIEWER: Was civil engineering different from some of the other engineering departments in that respect or were they all making that transition together?

ROOS: I think they were all making that transition together. Electrical Engineering led it, led it largely because Gordon Brown was department head of electrical engineering before he became dean of engineering.

But it was really a profound transition. One thing that helped it significantly was that we were in a growth mode in terms of hiring faculty at MIT. This is when the Ford Foundation, for example, was giving lots of money to MIT, so there was the opportunity to try new things, as opposed to the situation that we've had probably for at least the last 20 years, where we've essentially had no growth in the faculty. It becomes far more difficult, then, to bring about fundamental change.

INTERVIEWER: If this was the mid- or late-60s MIT, of course, had a president at that time who didn't have a doctorate either.

ROOS: That's true.

INTERVIEWER: Howard Johnson.

ROOS: Yes, that's true.

INTERVIEWER: I don't know if that made it seem a little more okay to have a department head who --

ROOS: I had never thought about that, but, yes, you're absolutely right. Yep.

INTERVIEWER: You mentioned your interest in transportation systems. What about that grabbed you? When did you start being interested in transportation and transportation systems, and what about it intrigued you?

ROOS: Well, I grew up in New York City, and so I experienced not only cars, but I experienced mass transit. And there seemed to me to be something very exciting about the growth of cities and urban areas.

And we were going through -- if you think back to the '50s and the '60s-- a fundamental transition. This was when all of a sudden suburbia was evolving and we were seeing fundamental shifts in terms of where people live and where people work. And, you know, in a sense transportation I viewed sort of as a window into understanding those phenomena, because there really was no factor that was stronger than the automobile in terms of its influence on the structure of cities, urban areas, and urban development.

So, that's what interested me. There were some books I read. Jane Jacobs wrote a very important book, and several others, that motivated me.

INTERVIEWER: But you didn't go into urban planning. You went into the engineering, although you took some courses, as you say. Do you think you would go into civil engineering today if you were an undergraduate now?

ROOS: No, I wouldn't. As a matter of fact, the Engineering Systems Division now is looking at implementation of an undergraduate program.

But the reason I didn't go into urban planning was that I've always loved math and I've always identified with being an engineer. I think there's another reason as well. And that is I've always been interested, in my career at MIT, in expanding the role of the engineer. In many ways, that's really what ESD represents. I think we're dealing today in society with very large-scale problems, very large-scale systems, and to be able to understand those systems one needs technical understanding. We're sitting here today with this horrible situation in terms of the oil spill, which, clearly, is fundamentally a technical problem, but it has so many other implications to it. And so, my feeling was that many of the problems that we're dealing with do not have an appropriate technical component to them. They don't have an engineer in a leadership position.

But for an engineer to take that leadership position, the engineer has to know more than the technical aspects of the problem. The engineer has to understand something about organizational principles, has to understand something about supply chains -- we're dealing with a global society -- has to understand the regulatory environment, and has to therefore understand something about the social sciences. And so that always intrigued me. That has been sort of a consistent motivating force throughout my career at MIT, even though my interest and my positions have changed.

INTERVIEWER: And the question then is how you cram that all into one undergraduate education without losing too much of your technical training?

ROOS: No, you can't. I think that it's really important that you have depth in some technical area, but you should at least have an understanding and appreciation in terms of the larger picture, the larger issue. And that's going to require going on and getting a graduate degree.

INTERVIEWER: You mentioned being interested in the transportation systems when you were growing up in the '50s, which was a period of real highway building, I think.

ROOS: Yes, it was.

INTERVIEWER: Around New York you look at --

ROOS: Robert Moses.

INTERVIEWER: -- the Long Island Expressway out to Long Island, which I think was problematic from the beginning. People said that it was a parking lot from day one.

ROOS: You know, it's very interesting you raise that, because Robert Moses, who -- I think he had about five different positions -- really ruled New York City and the New York metropolitan area. And I'm just sitting here thinking how I was a Brooklyn Dodgers fan, and the Dodgers had to move from Ebbets Field and wanted to move to a location within Brooklyn, which would have been a fantastic location, at Atlantic Avenue. And Robert Moses overruled it and said, no, you can't move there. You have to move to Queens. That's what motivated Walter O'Malley to go out to Los Angeles. And the Dodgers moved.

So, if Robert Moses didn't approve something, it didn't happen. And my father would come home from work really upset, because he did highway design and building. And he says, I just can't believe --

INTERVIEWER: In the New York City area?

ROOS: In the New York City area.

INTERVIEWER: Did he work under Moses?

ROOS: No, he didn't. He had his own consulting firm. But he said, this is just all wrong. This was the era of eminent domain, and basically, somebody could come in and just say, this is where the highway is going to go. I don't care what's going to happen to the neighborhoods.

And I just listened to him and I watched what was happening, and I said, on one hand, the use of the automobile and highways in terms of developing our suburban areas is fundamentally a good thing. People do want to have the opportunity to live in lower-density situations. But you have to do that in the right way.

And so, I think to a large extent, sort of watching that whole development take place and seeing, in a sense, the good side and the bad side was really what motivated me.

And then there were one or two faculty members who were in civil engineering who were just terrific.

INTERVIEWER: When you were an undergraduate.

ROOS: When I was an undergraduate. I took a course, as a matter of fact, as a sophomore, which was a design course. And that's what really convinced me to pursue my career.

And it turned out that they stayed around at MIT, and they were really the people responsible for bringing on a whole group of graduate students who undertook that transition I mentioned before, in terms of fundamentally changing the way you view transportation. And what we did at MIT had a worldwide effect on how the transportation field emerged. So it was really a very exciting time.

INTERVIEWER: And who were these people?

ROOS: Well, one of them was A. Sheffer Lang, who also turned out to become the first federal railway administrator when the Department of Transportation was set up. Bear in mind that there was no DOT at this point, that all the transportation was being done at HUD. And the second one was Marty Wohl. And both of them really played a role in revitalizing what we had at MIT.

INTERVIEWER: And one of them was your sophomore professor?

ROOS: One of them was my sophomore professor, and one of them --

INTERVIEWER: Which one was that?

ROOS: That was Shef Lang, and he also was my undergraduate thesis adviser, when I looked at the possibility of express buses replacing the Long Island Railroad. Turns out, it's very interesting -- this is a complete diversion -- the Long Island Railroad was not built initially to service the communities of Long Island. It turned out to be part of the most direct route from New York City to Boston, that you would take the Long Island Railroad to Montauk Point, which was on the North Shore. You would then take the ferry across to New London, Connecticut, and that's where you would hook up and continue on to Boston. So it turned out to be a very interesting thesis.

INTERVIEWER: Interesting.

What brought you to MIT? What made you focus on it? Tell me more about your childhood growing up in New York City?

ROOS: I grew up in Brooklyn.

INTERVIEWER: In Brooklyn.

ROOS: I grew up in Brooklyn.

INTERVIEWER: Only child?

ROOS: Pardon?

INTERVIEWER: Did you have siblings?

ROOS: Yes, I had two brothers. I was the oldest. And I always liked math and science, so there was certainly a motivation to go to MIT.

And, I'm Jewish, and that was a period where there were restrictions on admission of Jewish students to some of the best schools, particularly the Ivy League schools. So I went to a high school, which had 1,000 graduates, and we knew that every year you would have only one student at Harvard and one student at Princeton and one student at Yale, because that was the quota. But much to its credit, MIT had no quota. And so we had 13 students admitted to MIT.

I was admitted to Cornell and RPI, as well, but I came up to MIT, and I said, gee, I like this place and I like Boston, and so that's really what motivated me.

INTERVIEWER: Do you remember your first impressions? You say you came up and liked it. This was on a visit before you entered?

ROOS: Right, yes. Yes.

INTERVIEWER: And what did you see? Do you have any strong impressions that you recall?

ROOS: Well, I think in some ways, it was the Boston area that really became the decision-making variable. I looked at Cornell and I looked at RPI, both of which, obviously, were excellent schools. At that point, RPI was really pretty much comparable to MIT. And I just said, I don't want to be in this rural setting and I don't want to be in Utica. But Boston is really a wonderful place --

INTERVIEWER: Detroit, right.

ROOS: -- Detroit, that's right -- a wonderful place to be. I went to a few classes and I slept over, I think, in a dormitory and a fraternity.

INTERVIEWER: And here you came.

ROOS: Here I came.

INTERVIEWER: What were your undergraduate years like?

ROOS: Very tough. It's interesting, I think -- this is my 50th year graduating class -- and I got a chance to look at and talk to a lot of the graduates. And I was thinking of the comparison with the graduates today, and sitting and listening to the speeches at graduation.

When we were undergraduates here, about a third of the students flunked out. There was a famous story. Look to your left and look to your right -- this was the opening talk you got -- one of you won't be here for graduation. The motto was tech is hell. We all knew. We worked incredibly hard. We had classes on Saturday. We had compulsory ROTC.

And no one said to us, you're going to be the leaders of society, you're going to be entrepreneurs, you're going to create companies, which is essentially what today's undergraduates are being told. So in a sense we thrived on hard work, but it was really -- you know, the idea of the tech tool and the nerd. I think that that very much was the image, and a correct image, of what MIT was like.

But I think all of us felt we were striving for excellence. That, I think, was really overbearing. And it was also a fun place. You had a lot of really interesting people doing interesting sorts of things. So it was a pleasant experience.

INTERVIEWER: What did you do for fun?

ROOS: God, it's difficult to think back. But there were sort of hacks and pranks. I do remember -- this is when we still had a tremendous rivalry between the freshman and the sophomore class. The end of this thing was a glove fight.

INTERVIEWER: A field day.

ROOS: A field day. Yes, it's all gone now. I remember one prank where -- this was my freshman year -- we got together, and we managed to kidnap the president of the sophomore class, at Caltech.

INTERVIEWER: [LAUGHS]

ROOS: It's hard to remember this. And to get back, I remember, one night, some sophomores came into my room and they placed a hood over -- this is a true story -- placed a hood over my head. The next thing I knew, I was in a car being driven somewhere.

It turns out I was left in Providence, Rhode Island, with no clothes, just my underwear, in a fraternity house. So, it was that sort of stuff that was going on. In retrospect -- I hadn't thought about it in years -- that was sort of an extreme prank.

INTERVIEWER: Were you involved in politics, student government, or anything?

ROOS: No, no, no.

INTERVIEWER: Freshman Council?

ROOS: No, no. I don't know how I got involved in this.

INTERVIEWER: Interesting. Do you know whether a third of your freshman class didn't graduate? Or did it do somewhat better than that? Do you know how many classmates?

ROOS: Well, I think -- see, I think we wound up with 600 some-odd graduates. And we had started out with almost 1,000.

It was a large flunkout rate and quite a few left, also. We had exams every Friday in math, chemistry, and physics. Class average was always C, and it was not the grading system that one has today.

INTERVIEWER: Do you remember whether there was discussion while you were a student about whether the system should be changed? Because it was changed by 1966. The requirements had been cut in half and the flunkout rate went way down, I think.

ROOS: That's true, but I graduated in 1960.

INTERVIEWER: But I wondered whether discussion had started?

ROOS: No, no, I don't remember any of that discussion. The thing I remember most when you talk about the mid-60s-- I had become a faculty member in '66. One of the things I did -- This was actually before I became a faculty member. I was an instructor. After I got my master's, they made me an instructor, and I started what turned out to be the largest undergraduate computer course, Course 100.

And -- gee, it must have been '67 or '68 -- it was the Vietnam War. That I remember vividly in terms of what it caused at MIT. I remember troops marching down Mass Avenue. And here I was lecturing the largest undergraduate computer course. So I had several hundred students there, and I stood up in lecture and said, we can't talk about Fortran today. There are troops out here on Mass Avenue.

We did get into discussions in class about how the students felt. That was really a difficult period, and I'm sure many of us have talked about it who've been --

INTERVIEWER: Did you have a strong view on the war that you wanted to share with the students?

ROOS: No --

INTERVIEWER: Or did you just feel like you needed to --

ROOS: Yeah, I felt that --

INTERVIEWER: -- draw them out?

ROOS: Right. As a matter of fact, I thought it was really important that I take, more or less, a neutral position, so that no one felt they couldn't express their views.

But I did get drawn into it, in that -- You know, a lot of the issue at MIT was about divestiture of the instrumentation and the Lincoln Laboratory. And that was a time when Howard Johnson decided that, rather than divesting instrumentation lab, he would try and convert it to civilian application. And, in fact, asked Charlie Miller to become the director of the instrumentation lab. And at that point in time, Charlie, in addition now to being associate dean of engineering, had started something called the urban systems lab. I had the largest project in the urban systems lab. I had managed to get back into my transportation days. We had a large project with the Department of Transportation called Project CARS, Computer Aided Routing System, where we put a computer-based system in Rochester, New York, to be able to schedule, essentially, flexible buses around the Rochester area.

And so, Miller now was head, not only of urban systems, but head of the instrumentation lab. And he came to me, and said, I want your project to be sort of the big project for the instrumentation lab. So, let's get all the instrumentation lab people involved in Project CARS. And so, all of a sudden, I found myself somewhat involved in this issue of divestiture.

And sadly, a year later, Johnson and Doc Draper came to the conclusion this wasn't working, and therefore, instrumentation was spun off and became Draper laboratory.

INTERVIEWER: Were you conscious of it not working in that year leading up to that decision?

ROOS: Well, our project was working, Okay? That was working. But what wasn't working was the ability to bring in large-scale funding to be able to continue to fund all the other people in the instrumentation lab.

INTERVIEWER: To replace the defense department funding that was going to drop away?

ROOS: I think -- I don't know if this is correct or not -- but I was under the impression that Johnson had reached some agreement with Washington in terms of bringing in major funding, and for whatever reason it fell through. And when it fell through, it became clear that the funding just, unfortunately, wasn't --

INTERVIEWER: What happened to Rochester and CARS? How did that work?

ROOS: Well, we eventually did implement the system. We called it Dial-a-Bus. And it got me interested in a field of transportation that was largely overlooked, which we call paratransit. And paratransit was essentially what I would call informal transportation.

So, for example, just think of how many taxicabs there are in the United States, and the role that taxicabs play. And we started to get involved in special services for the elderly and the handicapped, like the RIDE service here. It really was the launching point for the first professional look at these types of services. Particularly if you look outside the United States, you have jitney por puesto service. And what we started to discover was that public transportation in most urban areas was highly regulated. You had an organization like the MBTA, which was unionized, which did not allow other competing services to exist within the metropolitan area.

And so we started to look at that question and how one could basically introduce more variety of services and, to some extent, deregulate what was going on in the urban area. And that got me interested to a large extent in institutional issues, the fact that you can't just take a technology and put it in. You have to worry about institutional reform, and that certainly was true in transportation.

INTERVIEWER: Come back to your Course 100 and the large-scale computer systems. What exactly were you doing and why was it in civil engineering?

ROOS: Interesting question.

At that point, essentially all of the educational aspects of computers -- and mostly research -- were, as you would expect, in electrical engineering. But their approach was almost entirely to train computer professionals. And so, their courses were, how do you build an operating system? And they weren't that much interested in the computer from an engineering problem-solving perspective. And I'm in no way saying that in a critical sense. There was a tremendous need for developing computer professionals at that point in time, both hardware and software.

During my PhD I took all those courses in electrical engineering, so I really did understand how to build operating systems and the hardware of computers. But Charlie and I recognized that there was a need for all these engineering students to understand the computer as a problem-solving -- not only a problem-solving -- tool, but fundamentally something that could change the whole way you looked at engineering problem-solving.

And so we introduced Course 100 as a course in engineering problem-solving. And students took the Fortran language, which was a language that lots of people look down on, and for good reason, in terms of its structure, but nevertheless was really important in terms of a tool for engineers to use both in research and in applications.

And so, that really started us on this path of a major focus -- both research and education -- on computer applications. We got very involved in Project MAC. Project MAC was the big time-sharing project, but it not only was to create a computer system that could serve multiple users, it was also to be able to show new approaches to software. That was very much when we started to get involved in ICES.

The civil engineering department wound up having the largest computer system on the MIT campus other than the computer at the MIT Computation Center. We had an IBM System/360 Model 40. So, we played a pretty major role.

INTERVIEWER: Did you have any conscious sense as an undergraduate, or even as a graduate student, that MIT was a place where you wanted to stay for your whole career, or that you might?

ROOS: As I said, I wasn't even aware that I wanted to be a faculty member. But once I was a faculty member, I was very happy here. I don't know if I made a conscious decision I was going to stay the rest of my life, but I certainly was going to stay as long as, first of all, MIT wanted me to stay. Of course, there was a tenure decision. And as long as I continued to be interested and excited, and that turned out to be the rest of my life.

INTERVIEWER: You mentioned that as a sophomore you seemed to have some relationship with at least one of your professors, Shef. How did that come about, and did most of your friends know professors? Was that a fairly common occurrence, that those kinds of relationships took place? Or was that unusual?

ROOS: I'm not sure. Let me answer the first part.

This was a design course and there were relatively few students in it, because there were relatively few students in civil engineering. And so, there was the opportunity for much more interaction with the professor than one would have, say, in a lecture or recitation course. And he turned out to be a very, very charismatic individual, and a very welcoming individual. So I think it was a bit unusual, Okay? Yes, I think it was.

INTERVIEWER: Because it seems to me that those types of relationships are in some way at the heart of education, and drawing students in, and so if one can look at what makes it happen when it does, maybe one learns something.

ROOS: I'm just reluctant to answer --

INTERVIEWER: Put too much?

ROOS: It was too long ago in terms of other people. But it certainly had a huge, huge impact on me. At that point in time, I think it's fair to say I was still somewhat undecided as to precisely what my major would be. I was inclined to transportation, but this was sort of the litmus test to see was that the right decision. And there was no question.

INTERVIEWER: Since you weren't thinking about academia, what were you thinking about doing after college? Did you have a vision at that point?

ROOS: No. I always liked to solve problems. I thought I would go out and get a job -- I interviewed at a variety of different companies -- and just see what things were like. And I had a feeling, at some point, I might want to come back to get an advanced degree.

It turned out that it was quite by accident that I wound up getting a research assistantship a week before I was going to leave MIT and get a job. That was extremely fortunate, to say the least.

But I don't think I've ever really planned where I was going to be in five or ten years, and as I look at my career, three or four things happened absolutely by accident, and really changed what I was going to do.

INTERVIEWER: As you look at your colleagues and your friends, do you think your career was unusual in that respect, or did that happen to many of them?

ROOS: Let me answer that in two ways, okay? There were a group of us in civil engineering who were brought in, all of whom I would consider non-conventional civil engineers. So they were purposely brought in by Miller and Brown to change the department.

All of us pursued leadership roles at MIT. So, for example, Fred Moavenzadeh has led numerous international programs, and Dave Marks headed the energy lab and was department head, and Richard de Neufville headed the Technology and Policy Program, and Joe Sussman became a department head. So, that group was all similar to me and very non-conventional.

But collectively, I'd say that group was very different than most of the faculty at MIT. We were all interested, I think, in fundamental change, not only in civil engineering, but of engineering in general, to make it, in a sense, more interdisciplinary, more socially relevant, more having engineers play leadership roles in terms of the major problems, the really great challenges that we face.

INTERVIEWER: As you kept getting these invitations to stay on and do more, particularly, say, when you took your first faculty job, did you stop and think about, is this a fork in the road that I really want to take? Am I giving something up by not going to industry, either financially or in terms of experience?

ROOS: No, I always worked with industry in every position I had. I also had the opportunity to start a consulting company, and so I felt like I got that sort of satisfaction, as well as what I was doing at MIT.

And one of the things I have always been interested in at MIT is the university industry connection. In fact, at one point -- I think it must've been maybe 15 years ago or so -- Glen Urban, who was then dean of Sloan, and I co-chaired a commission for the president on MIT university industry relations. It was a very interesting assignment. We had a great, great group. Lester Thurow and Mike Dertouzos -- oh, gosh -- Nick Negroponte, and we went out and we interviewed 15 CEOs to find out how they viewed MIT. So there's always been that industry interest.

INTERVIEWER: And what did you learn when you interviewed 15 CEOs?

ROOS: Well, I think we learned that it was really important to do voice of the customer interviews. That some of them had, first of all, some very good suggestions as to how they felt MIT could work better with industry. And some of them had some very legitimate concerns about MIT. So, we wrote those up.

INTERVIEWER: Do you remember any of them in particular?

ROOS: I think there were two, in particular. One, which continues, was the whole question of licensing and intellectual property. And I think that will always be an issue with industry.

The second was some of them felt that we were not paying enough attention to companies in Massachusetts and in the 128 area, and that we should do more. And I felt some of their complaints were quite legitimate. In fact, I wound up interviewing Lew Platt, who at that point was CEO of Hewlett Packard, and it turned out that prior to that he had run -- I think it was -- Hewlett Packard Medical Devices here in the Boston area. And he told me how, on a couple of occasions, he had wanted to do some things with MIT and it just didn't work out. He felt that MIT was missing an opportunity, and in listening to him, I agreed.

INTERVIEWER: This was when? '70s, early '80s, '90s? We can look it up.

ROOS: Twenty years ago maybe?

INTERVIEWER: Roughly?

ROOS: Yes.

INTERVIEWER: Interesting. And did MIT make many changes or any significant changes as a result, do you think?

ROOS: Well, this was a period where we were considering what more MIT should do with industry. And yes, as a matter of fact, now that I think of it, several important things came out of that study. Two or three, in particular. We started the Industrial Partnership Program as a result of that. And we had partnerships with Ford and Merrill Lynch and DuPont and several others.

Another very important thing was we called together six faculty entrepreneurs for dinner to talk about their experiences. And all of them had said that they didn't discuss their experiences, because they had made a lot of money and they felt that it just wasn't right to do it.

But on the other hand, they were missing an opportunity to bring this to the students. And also, all of them felt that at some point in time, they had had a real problem, and they didn't know where to go. And Dave Staelin was one of the faculty at that meeting and out of that came this Venture Mentoring Service idea, which has turned out to be a tremendous success. And we also talked about it as a way of potentially raising money for MIT, that if faculty and graduates were successful in starting new ventures, that hopefully they would give about 5 percent of their money back to MIT. So some very, very good things did come out of that commission.

INTERVIEWER: Who was the president who put this together at the front-end?

ROOS: Well, it was Chuck --

INTERVIEWER: It was Chuck, then, so it had to be --

ROOS: Yes, it was Chuck, and Joel Moses was Provost, so that was when it happened.

INTERVIEWER: Late '80s, early '90s?

ROOS: Yes, yes.

INTERVIEWER: Interesting.

So, let's go back before that to your auto projects and how you got into the big study of automobiles and what came out of that.

ROOS: Well, I had been appointed head of the Center for Transportation Studies, and I said -- [COUGHS] excuse me -- what project would bring together an interdisciplinary team at MIT? And I came up with the idea of the future of the automobile. And so, I went to Alan Altshuler, who was the head of political science, because I knew that it would be important to have a balance of engineering and the social sciences.

And I said, Alan, will you co-chair this with me? Alan had worked with me on the Project CARS. And Alan said, sure. But then we had to raise some money, which is always a little bit difficult. And I had gotten to know Ken Orski, who had been in charge of transportation at OECD. And he was just coming back to the US to be executive director of the German Marshall Fund. And so I went to Ken. I said, this is a good project. He said, I agree. German Marshall Fund will give you some money.

And then I went to Paul Gray, and I said to Paul, we've got this project and we've got base funding, and do you think we can get some more funding. And Paul thought it was a great project. The idea was that we would do some research, and we would also have every year some policy forums, where we would bring together from around the world leaders in transportation. Both from the auto industry and also from government and from labor.

And so Paul was able to get us some money from the Lilly Foundation. And we brought Purdue in because, Indiana. And so the project got started. And I still remember, we had our first meeting. Paul came to the meeting, and we had this at Eagle Lodge, in Philadelphia. And we had all these important auto executives. And this was just the time that Japan was starting to make significant inroads in terms of competition in the US. We had more or less lost some industries at that point to Japan, like consumer electronics, and now all of a sudden here was Toyota and Honda starting to sell lots of cars in the US.

And so we had this off-the-record policy meeting, and we heard all of these US executives saying, this is unacceptable, that we are having unfair competition from Japan. And Alan and I came away from this meeting saying, gee, we thought we had a project on transportation. This is not a project on transportation. This is a project on international competition. And so, all of a sudden, we really changed the focus of that project.

We went to Japan the next year, and we had a meeting, a policy forum. And prior to that, we sat down and we had discussions with the Japanese auto executives. And they said to us, you can have a policy forum here, but you can't talk about competition. That's off the agenda.

And so I ran these forums, and the way the forum worked was we had a square table, and people sat in alphabetical order, so we got away from issues of who sat where and who was more important. And the first thing we did was go around the table and ask people to introduce themselves. So the first person said, my name is Amaya. I'm a member of the Diet. And the last year I've spent negotiating the voluntary trade agreement between the US and Japan, which had been negotiated as a way to kind of hold down the battles that were going. And he said, I have a meeting at the Diet this afternoon, and obviously the issue that we need to discuss is international competition. And I'm sitting there saying, so much for Japan, Incorporated. Here is a Japanese saying we better discuss it, and that just opened the agenda up.

And we had the head of MITI there, and we had a senior executive from GM, and for the first time, there was this dialogue on what was happening in the auto industry. And the GM executive came up to me at the end of the meeting and he said to me, boy, this is worth our entire investment, the ability to talk to somebody in the Japanese government and to do it in an unscripted manner, to really discuss the issues.

MIT really was serving as an honest broker. So we brought to this meeting the results of our research, and we said to this group, did we get it right, what additional research should we do, and most importantly, what are the implications of this research? So they all talked to one another.

So we had four of these meetings, and they were all off the record, and out of that came our first book, The Future of the Automobile.

INTERVIEWER: What happened after that?

ROOS: The final meeting of the Future of the Automobile was at Kresge. That was an open meeting. We had Phil Caldwell, the CEO of Ford. And one of the people on the panel was Jay Chai, who was the CEO of the largest Japanese trading company in the US. He was married to the top Wall Street analyst of the auto sector, who had been part of our study. And they came up to me and they said, when you're in New York, stop by. We'll have dinner.

At that point, I had no intention of doing anything further. It was the end of the study. So about six months later, I had dinner with them, and they said, are you going to do a follow-on study? And I said, no. This was just too difficult to do, too difficult to raise the money, and if we did a follow-on study, we would have to raise all the money to make sure that it was done in the right way the first time we did it. Each country raised their own money and some of them did not follow instructions, so we were not able to do relative research.

And Jay said, how much do you need? And I said, we need at least a million dollars from the US, a million dollars from Europe, and a million dollars from Japan. He said, a million from Japan? I said, yes. He said, done! I said, done? Yes, he said. I will raise it, and if I can't raise it, I will give it to you, and so let's get on with it.

If we hadn't had that dinner, there never would have been an International Motor Vehicle Program and there never would have been the book The Machine That Changed the World. So we did another four-year study. We used the same format of the research in the policy forums, and out of that came The Machine That Changed the World, which basically said, in spite of what everyone is saying as to why the Japanese are so successful -- Iacocca was saying it was the yen dollar, and some were saying it was simply, Japanese workers had a culture that allowed them to produce cheaper and better cars. None of that was true. Basically, they had developed a superior manufacturing and production system in comparison with mass production, which we termed lean production.

And we benchmarked all of the manufacturing plants around the world and showed what dramatic differences there were, not only between the US and Japanese, but between the US and Europeans, as well.

So this program had a huge impact not only on the auto industry, because what we said was that lean production was not geographically dependent, that not only could you do it in Japan, but you could do it in the US. And in fact, this is when the first transplants were occurring in the US with productivity and quality to match the best in Japan.

INTERVIEWER: You mean, the Japanese companies were setting up production sites here?

ROOS: Exactly, which were called transplants. So you had Honda in Marysville, Ohio. You had the NUMMI plant in California. And secondly, we were saying, lean production is not only applicable in the auto sector, it's applicable in any sector. And so this became important from an industry point of view, and that's why we wound up selling 600,000 copies.

INTERVIEWER: Was this very different from the conclusions that you reached in the first study. And if you were using basically the same methods, why did you get further? Did it just take more time?

ROOS: The first study started to show that there were some really fundamental differences in the way the two countries were approaching things. But it was just the first look. It wasn't comprehensive. But I think, more importantly, it said there are really fundamental differences here in productivity and quality, but why are they occurring? What are the characteristics -- what are the differences of the system?

And when we looked at it, mass production and lean production are just fundamentally different in so many ways. The whole question of a focus on quality, the idea of basically constant learning and improvement, continuous improvement. The idea of different sorts of relationships with suppliers, working in teams. Just a whole different philosophy and structure between the two systems.

INTERVIEWER: The American auto companies had contributed to the second project?

ROOS: Yes.

INTERVIEWER: Were they at all angry when you came out with the results, or surprised? What was their reaction?

ROOS: It varied. Okay? And, not only the US companies, but worldwide. It varied.

Chrysler welcomed the study, and I think Chrysler welcomed the study because they were going through their third near-death experience. And they recognized that they had to fundamentally change. And I think when a company is in pretty desperate shape, that they tend to be more willing to accept things.

Let me give you an example of the different approaches -- or a different acceptance. I wound up giving a lot of talks to these companies. And so, there was one talk I was giving at Renault. And the CEO of Renault was Louis Schweitzer. Very, very smart guy. And I still remember Schweitzer getting up and saying, we have one of the authors of The Machine That Changed the World here today to describe what is in the book. And what he's going to describe is really important. It represents a totally different way of thinking about how you make automobiles. I want you to listen to what he has to say. But more than that, I want you to then think of how it's going to be implemented here in Renault, because your job depends on it. Okay. So that was one introduction.

The second introduction is at General Motors. Okay. And the CEO of General Motors stood up and said, Professor Roos is here to give you some academic perspective -- immediately sort of saying, academics, not industry -- on something they call lean production. It's very similar to what we've been doing here for the last 10 years, but we thought it would be interesting to have him come in and talk to you. Which to me said, don't pay very much attention, but we had him come in.

And General Motors was very upset with the study, very upset, because we used certain examples showing how badly they were doing things. And, by the way, all of what we did was publicly available.

We also had some huge, huge problems with Mercedes, very nasty problems, because, in the book, after we did a study of mass automobiles, somebody said, why don't you look at luxury cars? Which we did. And so in the book, we had a statement that it takes a certain European luxury car manufacturer more time to repair defects when the car comes off the assembly line than it takes a Japanese luxury car manufacturer to build the car to perfect quality on the production line.

And so, Mercedes immediately said -- we never mentioned names -- but they said, oh, well, they'll think of Mercedes. They'll think of Mercedes. And they said to us, you can't publish this book.

Now, bear in mind that we had sent copies of the book to every one of the companies we worked with to say, look it over, if there are mistakes, let us know, because we want to publish something that's correct. But we retain the right to publish. Mercedes never got back to us. They never came to our final meeting, when we discussed the chapters. The vice-president of research called me up. And I pointed out to him, I said, sir, you had three months to get back to us. And you basically didn't.

And the conversation got very, very nasty. And so at the end, he basically said to me, if this book is published, you will never work in Europe again. So we published the book. About a year later, Chuck Vest sends me an e-mail saying, I'm about to meet the president of Daimler-Benz, which of course Mercedes is part of. So I sent him an e-mail. I said, heads up, Chuck. You ought to be aware, and I told him what happened.

So Chuck goes to the meeting, walks into the chairman's office. The chairman is holding a copy of The Machine That Changed the World. Chuck says, oh, god. So the first thing is Reuter, who was the CEO, said, I want to thank you for publishing this book. It was an eye-opener to us. We never would have realized this, unless somebody from the outside had come in and pointed it out. And Chuck said, I'm delighted to hear you say that, because the Director of Research was really upset and told Professor Roos they shouldn't publish the book. To which Reuter said, you mean, the former head of Research and Development. We fired him. And furthermore, if you're going to do a follow-on study, we want to be the first company to sign up. So, to Mercedes' credit, they really did understand what was --

INTERVIEWER: And they made changes.

ROOS: And they made changes, absolutely.

INTERVIEWER: And in the US, to what extent -- you said Chrysler seemed pretty open, and GM seemed pretty closed. So, five years, ten years down the road, how much had they really changed?

ROOS: Well, you've seen what's happened to GM. And you had a company that had 50 percent of the auto market in the United States just sort of gradually lose that market, and of course, eventually, go bankrupt.

The sad thing was -- I mentioned before, one of the first transplants was the NUMMI plant. And this was a former GM plant in Sacramento, California, that had the poorest quality, poorest productivity, poor labor relations.

Joint venture with Toyota. Plant gets re-opened. 90 percent of the workers who are rehired are former GM workers, and they continue to work for the UAW. It's a Brownfield plant. It goes from having the poorest productivity, quality, worker relationships, to having the best productivity, quality, and working relationships of any GM plant in the country. What happened?

Toyota brought in the senior management, implemented lean production, the union cooperated. They went from several hundred job classifications to three. Workers formed teams. Focus on quality.

So here's a model of what GM should do. You know, in many ways, GM was embarrassed by it? This was Roger Smith putting in billions of dollars trying to automate GM into the 21st century, and here is this little experiment that's showing how to do it. And they just never wanted to accept it, and when they accepted it, they really didn't understand how to take it and be able to move it to the other GM plants. They would send somebody in for two or three days from a plant, saying, look at it, go back, and do it at your plant.

Well, by God, that's not the way you do it. I mean it takes six months or so. So they just never were able to fully understand and appreciate. And there was a culture there that it was literally impossible to turn around.

INTERVIEWER: Where was the UAW on this?

ROOS: Mixed. There was a man in the UAW, who was a member of our group, Don Ephlin, who was just absolutely terrific. Ephlin was largely responsible for NUMMI, and then Ephlin was the person responsible for Saturn, which you might recall was this very innovative company.

Ephlin ran for -- Ephlin was head of the GM unit of the UAW -- he ran for being the head of the UAW and he lost to a very reactionary guy, a guy by the name of Yokich, who basically wanted to undo everything that Ephlin did, even going to the extent that -- Where they made Saturn, there was a street called Don Ephlin Way. And he insisted that the street be renamed. So it just really set the UAW back.

More recently, now, with all the crises going on, I think the UAW has recognized that there's a need to innovate, and the current head of the UAW has been much more innovative.

INTERVIEWER: Have the transplants continued to thrive? The Japanese companies in America? Or have they --

ROOS: There are two auto industries in this country. There are the big three, Detroit-based, unionized, and then, throughout the South, every Japanese auto company and most European auto companies now. You've got BMW, Mercedes, Volkswagen. All have plants in the United States, all nonunion plants. So it's essentially, really, two auto industries.

INTERVIEWER: To what extent do you think the American companies could have prevented the crisis that occurred in recent years? Some of it seems to be a question of just having too many retirees with very generous benefits on the base of many fewer workers. Is there any arrangement under which that would have worked out, without having a pretty big amount of dollars per car as sort of a drag on the saleability?

ROOS: Well, there's no question that that has a huge impact, particularly now, when the number of retirees has increased and the sales have gone down. But if you go back 10 years, or even further, there were lots of things that the US companies could have done to prevent what's happened today. Every one of them made absolutely fundamental mistakes. They diversified into the wrong industries. They turned out bad product. They went into ridiculous joint ventures. Every one of them, chapter and verse, of a huge number of mistakes. A terribly inbred industry that just really didn't understand how to transform itself. What happened did not have to happen.

INTERVIEWER: Professionally, you moved into other broader public policy issues after that. You were a member of the MIT Commission on Industrial Productivity, and then some other commissions. Talk about that shift and some of what you did next.

ROOS: Well, part of it was accidental. I mentioned before, several things happened in my career that were unanticipated.

I was heading the Center for Transportation Studies. And there was some study, I know, going on at MIT that was looking at what MIT you should do in the policy area. And I got a call -- I think it was from the dean of engineering, I'm not really sure -- saying that this commission has recommended that a new center be set up at MIT, a new policy center, and you should head it. I said, what? I'm not a policy person. I have never taken policy courses. And they said, well, we've seen what you did on the auto study, and what you did on the Center for Transportation Studies, and you work with social scientists, and so we think you would be an appropriate person.

So I agreed to do it. We called it the Center for Technology Policy and Industrial Development. One of the first things I did was to ask two colleagues, one from Sloan and one from political science and economics, to be associate directors. It was Charlie Fine and Mike Piore, so that it really would be very much of an interdepartmental approach. And that did get us into a number of areas where we looked at some significant issues. Particularly environmental issues. We started with a major program on hazardous waste, and this was really the beginning of the whole issue of sustainability.

And so this center became, to a large extent, the focal point for working on problems of sustainability. John Sununu, who has a PhD from MIT and was dean of engineering at Tufts, and at that point was Governor of New Hampshire, came down to help us as a political person, and getting involved in some of these policy issues. And that worked out very nicely.

We also had an advisory board. And the advisory board was really quite good. And they said to me, you know, we really like what's happening here, but there should be more than a center. We really ought to have an academic unit at MIT that brings together engineering and social sciences and management. And so, in a sense, that was really the beginning of the discussions of what turned out to be ESD, Engineering Systems Division, following, I think, at least 10 attempts to set up a similar type of unit at MIT, all of them unsuccessful for various reasons.

INTERVIEWER: Had you been involved with any of the earlier ones?

ROOS: One, just one.

INTERVIEWER: And why do you think they were unsuccessful, and why did this get off the ground --

ROOS: [COUGHS] Excuse me. INTERVIEWER: -- what did it take to get it off the ground?

ROOS: Well, I think there were basically two reasons. The first reason was that there were a fair number of people who were unconvinced that this was an appropriate thing for MIT to do. People in the Engineering School in particular. That it was better left to political science or management, rather than forming an integrative unit with them. These were people who believed if you didn't have an equation and you didn't have mathematics, it wasn't legitimate, that it was all very fuzzy stuff. And so there was a lot of questioning in that direction.

I think that the second was university politics, and simply the zero-sum issue of, you have a certain amount of money, and you create a new unit, and it means in all likelihood, there's going to be less money for the other unit. And so, as I say, each one of them was somewhat different. In one case, it was really the Humanities School that objected. This was a proposal when Nan Friedlander, and Gerry Wilson, and John Deutch were the people heading humanities, engineering, and provost, wanted to basically take STS and combine it with this center and broaden STS.

INTERVIEWER: Which was science, technology, and society?

ROOS: Science, technology, and society, yes. And the faculty in science, technology, and society objected to it. And so that never went forward. And there were several others, as well.

INTERVIEWER: So talk about the Engineering Systems Division, which can now give degrees of its own?

ROOS: Absolutely.

INTERVIEWER: And how much is it a technical engineering degree, and how much isn't it? What kind of mix did you end up with?

ROOS: Well, let me take a step back, because there was an important motivation for setting up the Engineering Systems Division which relates to degrees.

We had seen a number of new degrees established. For example, the technology and policy degree. The leaders for manufacturing degree. The system design and management degree. A degree in supply chain and logistics.

INTERVIEWER: Were these undergraduate degrees?

ROOS: No, these were all graduate degrees. However, none of these degrees had the ability to hire faculty. And in fact, if you took a degree in Technology and Policy, you had to be admitted through a department. In other words, none of these could directly admit, because only a department at MIT could admit students and grant a degree.

So we had kind of a weird situation, where it was somewhat dependent on the view of the department head as to whether somebody was appropriate for technology and policy. And so one of the main motivations for setting up the engineering systems degree was to provide legitimacy, through all of these degrees that already existed, to be able to directly admit their own students, for us to be able to appoint faculty. There were no faculty in technology and policy, because it had to be voluntary, in addition to your departmental responsibilities.

So Engineering Systems Division, to a large extent, the primary justification was educational. However, in the view of the faculty who established it, it went far, far further. You know, if you look today at the grand challenges that have been compiled by the national academy, these are all large-scale system problems. By the way, many of them -- if you look at the grand challenges of the 21st century -- many, if not most of them, are to correct problems created by the grand challenges of the 20th century. We have problems in terms of global warming. We have problems in terms of adequate water.

And so it becomes clear that the really big challenges require, A, an understanding of the behavior of large-scale systems -- large-scale complex systems -- and B, require an interdisciplinary approach. And that's really what the Engineering Systems Division is doing.

We had a committee formed when we decided that there ought to be a unit. And the reason that we decided to call it a division rather than a department was we wanted it to be integrative. So everybody in the Engineering Systems Division has part of their appointment there and part of their appointment elsewhere, and not just in engineering. Right now, there are faculty appointments in every one of the other schools at MIT. Most in management, but also in science, in humanities, and in urban studies. And we also created a new designation for faculty, which was the dual appointment. Dual is equal-equal between the two departments.

INTERVIEWER: Which had never existed?

ROOS: Never existed, no.

INTERVIEWER: Did that take a faculty vote or anything?

ROOS: No. I'll tell you an interesting story. As a matter of fact. Talk about faculty votes.

We went through a painstaking process to create the Engineering Systems Division. We went before every academic committee. We went before the Committee on Curricula and had to go before the faculty. It turns out, in meeting with Chuck Vest initially, Chuck said, you know, we could create the Engineering Systems Division in five minutes. I said, what? He said, the Corporation has the sole responsibility at MIT to create departments and other units that grant degrees. The faculty has no say in it whatsoever. So the Executive Committee of the Corporation could've simply said, yes, we approve. Engineering Systems Division goes to a vote of the Corporation.

Chuck said, don't do it. You'll have no legitimacy, no buy-in by the faculty. So we did go through that entire process. We managed to turn around Engineering Council, which initially opposed it, and eventually got a unanimous vote from Engineering Council to create it.

INTERVIEWER: How did that happen? Was it a turnover in who was sitting there? Or did you have such persuasive arguments?

ROOS: There were several reasons. The first, was that I was appointed associate dean, and so I had a year working with the Committee of Engineering Council to look at the issues. And so I had some allies, as a result of that, who became convinced.

Secondly, at the crucial vote, Ed Crawley was then the head of Aero and Astro. And Ed was also teaching a course in system architecture in the System Design and Management Program. And Ed said, you know, I was very, very skeptical about this, and I got involved in teaching this course. And if I were to look at this as a field, I would say, it's maybe 20 percent developed at this point. So one way of looking at it is to say, wait a second. It's 20 percent developed. It's not ready. You know, let's do a little bit more before we are really sure this is something MIT should do.

He said, on the other hand, it's exactly what MIT should do! We should be the people developing the other 80 percent, because this is really important stuff. And I think having Ed as a department head supporting me was really very, very convincing. And so it was a unanimous vote.

INTERVIEWER: And when he called it a field, does that mean a new discipline with new techniques? What does that mean? In other words, that he was convinced it was a new body of knowledge and skills, but they just hadn't all been formulated yet?

ROOS: Well, you see, I think it's important to use the word field, rather than discipline. In fact, it's really multidisciplinary. Yes.

Are there basic principles, underlying principles which define it, and are there areas that constitute fields of study? Basically, can we define a PhD for this? And that's been the primary objective of the Engineering Systems Division over the last 10 years.

INTERVIEWER: And will they get hired, I guess would be the other question.

ROOS: Will the faculty -- the PhDs?

INTERVIEWER: In other words, you can define a set of studies, but will anybody want these people?

ROOS: Well, the answer is yes. Our graduates are in great demand. The first thing I should mention is that we essentially have 100 percent acceptance of people who we accept for the PhD. So people see this program as really important.

INTERVIEWER: Which is about how many a year?

ROOS: It's about 15. We send our graduates to different places. We send quite a few of our graduates to engineering programs -- to engineering system-like programs -- throughout the world. We send some of them to management schools.

You mentioned in your introduction that I started this organization of engineering systems universities. We have 50 members now. And what's interesting is that they all have somewhat different names. In a sense there's a branding problem. In Carnegie Mellon it's Engineering and Public Policy. In Stanford it's Management Science and Engineering. In some schools it's Industrial and System Engineering. In others it's Enterprise Engineering. In others it's Macro-Engineering. All of them are engineering-school based and they're all taking this somewhat broader perspective towards engineering.

INTERVIEWER: Are any of them really different? In other words, how much is really a question of branding and how much is a question of content and product?

ROOS: The organization has existed for six years. And what I find interesting is that in the six years, there's been a convergence. That each of the programs has broadened and in broadening, collectively they are converging, in terms of looking at common information.

The second thing that's happened is there's really been a dramatic shift in the application area focus. To a large extent, most of these programs started out with a manufacturing base, looking primarily at autos, and at air, and at DoD.

INTERVIEWER: As at MIT.

ROOS: Yes. What's happened is there has been broadening. Broadening in two respects. First of all, much more interest now in computers and communication information systems, for obvious reasons. But secondly, much more interest in energy and health care. So I think that, yes, you see some differences, and in a sense there are really four areas that are a part of this.

One area I would identify as operations research/systems analysis. Another area would be system engineering/industrial engineering. Another area would be technology and policy. And the fourth would be organizational factors. In a sense, one could argue that engineering systems really comprises some of all of those things. You look at the field of management, for example, and there are very disparate units in management. The same is true of engineering systems. So schools place different emphasis, but I think all of them are looking at the same sort of issues.

INTERVIEWER: You say that there's been a broadening of many of them. Has that been true of the MIT program, as well?

ROOS: Well, the MIT Program is the broadest. And in some sense, I think maybe we're moving in the other direction, in terms of focusing more in a few areas. And so, yes, we're still very broad, but we have, over the past several years, chosen three or four areas to focus in.

INTERVIEWER: And are there undergraduates earning their only degrees in this area yet?

ROOS: No, there aren't, but we are seriously looking at the idea of a dual degree in engineering systems. The concepts of engineering systems we feel are really important to introduce into the undergraduate program, even if there isn't a degree in engineering systems. The dean of undergraduate education right now is Dan Hastings. And Dan succeeded me as the head of the Engineering Systems Division, so you have somebody there who certainly understands the importance and need.

INTERVIEWER: Another thread in your career has to do with international projects. You certainly had a big international component in the auto study.

You helped shape the Cambridge University- MIT partnership and you currently direct MIT- Portugal. Can you tell us about that those? One or the other or both?

ROOS: Yes, but let me preface it first by saying that we are living in a global society, and therefore, I think it is really essential for MIT faculty and students to be aware of international implications. That's one reason that I've gotten involved.

One thing that's striking, I think, about both of these relationships -- I wound up playing probably the primary role in both negotiations. And in the case of Cambridge, the negotiation was with Gordon Brown, who was then Chancellor of the Exchequer.

INTERVIEWER: A different Gordon Brown from the --

ROOS: No, same Gordon --

INTERVIEWER: Same Gordon?

ROOS: Oh, yeah, yeah --

INTERVIEWER: Who had been at MIT?

ROOS: No, no, no, no, no, he was Chancellor of the Exchequer in the British government --

INTERVIEWER: Right, but wasn't there a Gordon Brown at MIT that we talked -- that's what I meant.

ROOS: Pardon me, yes, a totally different Gordon Brown. Thank you, I hadn't thought about that. Right.

And with the Prime Minister of Portugal. The reason I mention that is that typical university contracts, say with NSF or with DOT, would be with some unit there. And maybe you would be talking of a million dollars or a few million dollars. All of a sudden now the negotiation is taking place at a totally different level, where the amount is more like 100 million dollars, which, of course, to a university is huge. To a Gordon Brown is a small part. But the fact that it's occurring at the level of a prime minister or somebody at a cabinet-level position implies that there are different expectations for the relationship. So in almost all cases, what they're interested in is the MIT model and the effect that it's had in terms of economic development, entrepreneurship, university industry relations. And so in both cases, that has really been a fundamental driver.

They're different, because Portugal is probably the poorest country in Europe, and therefore its needs were very, very different than the UK. Portugal has, to a large extent, lost its manufacturing base to Eastern Europe and to China. It realizes it has to move up the value scale. It realizes it has to get involved in knowledge-based industries.

So the Prime Minister came to office with his political platform based on science, technology, R&D, and higher education. And recognizing the need to partner with a Western university like MIT. And the focus, therefore, in Portugal becomes human resource development. One of the things we're doing is putting in place new PhD programs so that they can attract the best and brightest. University reform. University industry relations. It's a national program. We're working with all the universities.

And people at MIT said, well, why are you working with Portugal? You know, there are Singapore and Abu Dhabi, and very, very rich places that have far more funding. The point I make is there are three or four criteria, I think, to a successful international relationship. One is, do you have high-level political support? Absolutely, had it. Portugal. Second, do you have good colleagues? But thirdly, what is the impact? And if we could have an impact in Portugal, as I believe we will, it sends a really important message to all the other Portugals of the world, to say, it's really important to invest in science, technology, R&D, and higher education. So, yes, it's more difficult than working in a Singapore, but it's been a very rewarding experience.

INTERVIEWER: So, what exactly does MIT do in a relationship like that? Do one or two faculty members go and talk to them? I don't have a sense of --

ROOS: Sure. We have almost 50 faculty working on this program. The theme is engineering systems. We're working in four areas, in energy, in transportation, bio-engineering, manufacturing and advanced design. In each area we're developing a new PhD program -- very unique PhD program -- and an executive master's program. Faculty help developing curriculum. They go there for limited periods of time, a week or two, to lecture. We have Portuguese faculty and students visit us here at MIT. And we carry out an extensive research program in each one of these areas. And so, not only do we have faculty, but lots of graduate and undergraduate students.

INTERVIEWER: What do you think is MIT's role and responsibility in the world? And do you think it's changed much in the 50 years or so you've been a part of it.

ROOS: Well, I think its primary role is to educate the best and the brightest. That's the fundamental role. And to do first-class research. And secondarily, to be a public service. Has it changed in 50 years? In some ways, no. 50 years ago, I think the answer would have been the same. But in other ways I think it has, in at least two respects. 50 years ago, I think there was no question that MIT was viewed as, if not the best, one of the best universities in science and engineering. Okay. Now, MIT is viewed, not only as one of the best universities in science and engineering, but one of the best universities, if you look at US News & World Report.

So the agenda, in some ways, I think has broadened. The student body has broadened. Secondly, is the respect that MIT has and the fact that it's got a global reach. In some ways, that's both an advantage and it also is potentially a really tricky issue. The extent to which -- Is MIT a global university or is MIT a US. university with global interests, in terms of where the emphasis gets placed? What responsibilities do we have to the US above and beyond those globally? And that's a little bit of a tricky balance.

INTERVIEWER: Lots of room to keep inventing yourself every seven years.

ROOS: Right, absolutely.

INTERVIEWER: Thank you very much for your time and for your conversation and insights. Enjoyed it.

ROOS: Enjoyed it very much, as well. Thank you.