John A. Armstrong, “Is Basic Research a Luxury Our Society Can No Longer Afford?” Compton Lecture (1 of 3)

Search transcript...

MILLIARD: Anyway, John Armstrong is the 1993-94 Karl Compton lecturer. John actually, whom I've known for a very long time, began his career here in Cambridge-- sadly not here at MIT, but down the street at Harvard-- where he was a graduate student.

And I knew that he worked for George Benedek-- sitting over here-- and therefore got a very auspicious beginning. What I didn't know until about five minutes ago was that he was actually George's first graduate student. And presumably, this is why George has gone on to such a brilliant career.


And George's second graduate student was [INAUDIBLE], is a name you'll also recognize. So George started out [INAUDIBLE].

John then went on to-- to then doing a [INAUDIBLE] residence and switched into optics, where, as I understand it, he worked for [INAUDIBLE] actually, Bloembergen, and [INAUDIBLE] Harvard, and then from there went to IBM, where he had, first of all, a distinguished career as a scientist in the area of optics, and then has gone on to a distinguished career first as a manager of science and then as a manager of science and technology, finishing his career as vice president for science and technology at IBM Corporation [INAUDIBLE].

I have one personal memory of John, which I particularly appreciate, especially in my role as dean here, I try to remember occasionally or often, which is that-- I don't know if John even remembers. But in the late 70s, when I was still a young faculty member-- relatively young anyway.

And there was a young scientist at IBM by the name of Paul Horn, who sadly also is a high-up executive now. And we decided we wanted to start a research program in the area of [INAUDIBLE] radiation. And John was at IBM. And I was at MIT. And we decided that we needed some suitable capital for that.

So John went to the relevant managers at IBM at that time. I went to our post [INAUDIBLE]. And Paul went to John. And the requisite funds became available. Then we decided that we needed some sort of formal agreement. And so I remember going down to IBM and meeting with John, discussing [INAUDIBLE] get the MIT lawyers, ands the IBM lawyers, and the MIT deans, the IBM VP's together, and make up this elaborate institutional agreement.

And John said, whatever you do, don't do that. He said, keep all of the administrators out of it. And let the scientists take care of it. And it'll be fine. And so we actually did just that. And we managed a very successful long-term collaboration between my group and others here at MIT and IBM with no administrator ever almost knowing anything about the existence of the program.

So I try to remember that wisdom, that deans, et cetera, should just keep out of affairs. The only sad thing about that is that our research was sufficiently successful that all of the participants are now doing administration.


Anyway, so we're really fortunate to have John here. Just a little bit of history, some of you may not know just how distinguished this series is. The first Taylor Compton lecturer was actually Niels Bohr. And another lecturer in a somewhat different tradition, but equally distinguished, was none other than Hubert Humphrey-- who some of you may remember.


So finally, then John, as I said, just stepped down this year as VP at IBM. And we're very fortunate to have with him a residence here at MIT. And he's already been talking with many in our community. I believe he will be available to talk to others. Through his experience first in academia, and then in industry, and then in a lot of government service, has a remarkable amount of experience and wisdom he's going to share with us.

He also has some ideas which you may find provocative. And one thing about John which I admire is, if he thinks something, he doesn't mind saying it. And he is also willing to you disagree with him, which I hope some of you will at the end of the lecture today. So he's going to give us a series of three lectures, the first of which has the title, Is Basic Research A Luxury Our Society Can No Longer Afford.


ARMSTRONG: Thank you very much, Bob, for that introduction. I do indeed remember that [INAUDIBLE]. I appreciated your words of introduction. And it may be that George Benedek has gone on to great things because I was his first student. But then I've worried also that he may have left Harvard and come to MIT because I was his first student.


MILLIARD: Both two.

ARMSTRONG: Both may be possible, I suppose.

Well, I am sensible of the honor that it is to be a Karl Compton lecturer. And I want to publicly express my thanks to the president [INAUDIBLE] for this appointment, and to the colleagues here at MIT for hospitality and interactions.

As Bob said, one of the reasons that I want to give this talk this afternoon, which will be in some respects I think provocative, is that I do want to have your reactions and the critique. And I proposed to leave time at the end of the hour for that. But I would also welcome you hunting me up or down, as the case may be, in my office, which is 6-208.

Like many of you here this afternoon, I'm sure I'm thinking hard about the changes in the climate for research and development that have occurred in our country-- changes that are due to the end of the Cold War and the change in the support of research by the military; changes due to the declining competitiveness of some industries in our country; and changes due to the deteriorating fiscal situation in our country.

In a recent lecture at UC Santa Barbara about two years ago, I dealt with the issues raised for university research by too great a rush to tie that research to industrial competitiveness. The lecture was entitled Research and Competitiveness-- Problems of a New Rationale.

That lecture, however, did not have occasion to address in depth the issues which surround the definition of and the rationale for the support of basic research. And it's to those questions that I propose to turn today. Some of what I have to say may be disturbing. But I want to assure you, as I already have, that I welcome discussion, and that I do not think for a minute that I have the last word on these matters.

I have been trying to think through a rationale for society's support for basic research without leaning on the emotionally-invoked mantras or the naive analysis of impact that one sometimes hears from the science side, and without giving comfort to the impatience and the antipathy that sometimes characterize government officials and corporate executives.

One could hope to produce a rationale for basic research that is neither excessively immodest nor excessively modest in its claims-- a rationale which can be used in Washington and in corporate boardrooms, as well as at the NSF and in the faculty councils of our major universities.

One could hope for a rationale, which does not distort the complex history of science and technology as they have actually interacted over the last 150 years. Such irrational, if it could be found, might be proof against the present tendency to direct all research investments to the most immediate of political and social concerns, while at the same time giving these concerns the substantial weight they deserve.

You will think perhaps this is too grand a set of intentions. But I'm going to risk the attempt. The worst that can happen is that you send me back to my tractor on my farm in upstate New York.


Experience teaches us that it pays to listen carefully to our critics. And one of the critics we should be listening to is Representative Georgie Brown Jr, who is one of the all-time best supporters of science and technology, and who is the chairman of the Committee on Science, Space, and Technology of the US House of Representatives.

In an article in Science Magazine last October a year ago, an article called "Rational Science, Irrational Policy," Mr. brown had many thoughtful things to say about the national mood respecting public support for basic research and about the behavior of both scientists and public officials. But it was on the topic of basic research that I wish to quote him this afternoon.

"A more precise model," he wrote, "of the role of research in our culture might portraying applied research and development as sources of the technological innovation that fuel economic growth, ease the struggle for survival, and free increasing numbers of human beings to pursue self-realization through endeavors of the intellect and spirit.

Such endeavors," he writes, "include the search for new scientific knowledge through basic research as well as the quest for enlightenment through study and the practice of religion, philosophy, and the arts. Particle accelerators, spacecraft, cathedrals, and libraries are all essentially similar they are settings for cultural experience." End of the quote from Congressman Brown.

In other words, the congressman-- and he, as you know, is one of our friends. The congressman is saying that basic research is a luxury of individual self-realization that's getting too expensive. He does not seem to be saying that basic research is a necessity for modern industrial societies.

Thoughtful as I find the article as a whole, the comments on basic research disturb me. Knowing Congressman Brown, that may be why he wrote it-- not to disturb me personally, but to disturb all of us collectively.

In my view, he characterizes basic research by one of its minor attributes, namely the deep satisfaction sometimes expressed by well-known scientists that accompanied their important new insights into nature. Furthermore, the congressman's definition implies much too sharp a distinction between basic and applied research.

In my experience, first as a doer of research and then as a director of research, it was often hard to apply such sharp distinctions constructively, even to the programs or to the work of individual scientists. It's a distinction which, in its sharpness, confuses rather than clarifies public understanding and debate.

Now, I'm convinced that if we want to carry the day as those of us who believe in basic research-- and despite the provocative title, I will tell you right now that I do. You'll form your own opinion when I'm through.

But for those of us who believe in basic research, if we're to carry the day against those who want to turn all serious investment to the short term, we need to examine, among other things, our own culture, our own hangups-- if I may use that term-- and motivations, and our own often unexamined notions of basic research. The very concept is under a cloud today in many circles. We need to dispel that cloud if we are to answer that basic research is a necessity, not a luxury.

We also need to clear away some semantic underbrush from the term "strategic research." But whether or not basic research is a luxury, and if not, how much of it our society can afford are two different questions. And we'll have to keep both in mind as we proceed this afternoon. And I will try to deal with both.

Now, one of the things George taught me was to put the most important things up front. Before I give you the argument, I'm going to summarize it for you in a set of intentionally oversimplified assertions.

First, technical sophistication is a necessary, but not sufficient, condition for a modern, developed society to achieve its goals. Second, a vigorous set of research institutions is a key ingredient in many fields for helping a society to achieve technical sophistication. Third, world leadership in basic research is neither necessary nor sufficient for the achievement of strong economic growth and other national goals, at least over periods of many decades. A hush comes over the room.


Just so that you know what I'm thinking about, I'll expand on that in just a minute. The US became the country with the highest rate of productivity growth and the highest absolute productivity in the world in the latter half of the 19th century and early 20th century, long before we became the world leaders in basic research.

Japan has done likewise with respect to productivity growth since World War II without having yet become world-class in most areas of basic research. There are other examples, but they show that world leadership in basic research is not necessary for a society to achieve important goals.

That it is not sufficient is almost too obvious to require discussion. But the example of Great Britain after World War II, or the US semiconductor industry in the 80s, or IBM in the recent past should suffice to show that excellence in research is far from sufficient for the attainment of national or corporate goals.

Fourth assertion-- and this is the main point-- world leadership in basic research will constitute a competitive advantage if, but only if, it is coupled with world-class performance in the much more extensive and diverse set of skills, institutions, and investments that come between important scientific insights, and the creation of economic wealth, and the rising standard of living.

Whether that comparative advantage for leadership and basic research is a modest or a substantial advantage will vary by field and with time. But overall, it will be a comparative advantage we would be foolish to forgo in a world where the economic competition is getting fiercer and fiercer all the time.

Assertion number five-- finally, I will arguing that public support for the goal of leadership and basic research and of people important support for a rational strategy for reaching that goal will be achievable if and only if there are substantial changes in attitude within parts of the scientific community as well as changes elsewhere in government circles, in Congress, and in the public.

Now, I propose to develop these propositions in more detail. But first, I ask your indulgence while I say a word about my view of nature. Since I find on reflection that the various definitions of "research," of what is fundamental or basic, why it is important all tend to derive from the views held by the speaker about the nature of nature, if I may say so. And since these are seldom made explicit, they are often the cause of confusion in otherwise constructive discussions.

For my part, I am persuaded that God did not make the world according to the funding categories of the National Science Foundation--


--or according to the departmental structure of US research universities or even according to the priorities of the current Congress and administration. Nature is not divided up into compartments, like a wall of post office boxes. Despite what you might conclude from most course caveats, nature is interconnected and complex in ways that we still only dimly perceive.

Furthermore, in my view, nature is inexhaustible, or better said, quoting Freeman Dyson quoting [INAUDIBLE], "nature is infinite in all directions." That nature is inexhaustible or infinite in all directions leads to the possibility of a special type of exponential growth.

The possibilities for new scientific discovery will be proportional to the number of scientists working and to the budgets available. This, of course, is behind the often quoted factoid that more than half of the scientists who have ever lived are alive and working today.

But the consequences of this continuing exponential growth, however, are forcing themselves on us in new ways in the 1990s. As I say, they are a direct consequence of the nature of nature and its inexhaustibly.

These views have important bearing on the analysis, or least I claim they have important bearing on the analysis, I'm about to give. Because nature is inexhaustible, it is highly unlikely that all results of basic research will be of potential value. It's like saying all the numbers between 0 and 1 are of equal interest to mathematicians.

Because the set of unanswered questions is everywhere dense, we must constantly be on guard against the idea and the promise that if we just had more money and people, we would, quote, "get to the bottom of things." What is guaranteed to happen is that the enterprise will grow exponentially at a rate that need not have anything to do with society's needs.

Yes, the set of unanswered questions is everywhere dense in my view, but that is not any justification for giving up in our quest to understand nature or for a disparity of utility for mankind of much of what we do. It does, however, tend to put us in a somewhat different perspective.

The next thing I want to do is give you my definition of basic research. As background, I could, but will not, for lack of time, describe what I found by listening and observation are the interesting set of implicit, not-thought-out definitions of basic research used by many scientists.

Perhaps you'd like to return to that in the discussion period. But I choose to define basic research by what it is, not by what it is not. Moreover, I choose to place emphasis at least as much on results as on process.

Basic research, in my view, is research that leads to new understanding-- that's the key word, "understanding--" of how nature works, how things are interconnected, and what are the explanations of things. But applied research on the other hand may usefully be defined as research aimed at making something work.

This definition of applied research helps us to see how easily the line between basic and applied research blurs since it is an often useful strategy in making something work, to understand some aspect of nature's working. And conversely, many scientists who achieve new understanding in their basic research are highly motivated to turn that understanding into some new device, some new-made thing.

Incidentally, I do not favor the term "curiosity-driven" as a synonym for basic research or even as a part of the typology of basic research. First, it conveys an air of frivolity to too many taxpayers and congressmen. But more importantly, in my view, it is not as near the mark as the phrase "understanding-driven."

Now, from this vantage point, it's easy to be clear in my view why basic research is important to a country or to a company-- why, in fact, basic research is not a luxury. The importance of basic research stems first from the fact that, from time to time, it discovers or creates vast new possibilities for achieving society's goals.

It is also of value because, much more frequently, the smaller understanding is achieved allow advances or improvements in existing technologies, methods of health care, productivity, and benign treatment of the environment. Of course, having said that, we still have not yet dealt with how much basic research society should afford itself.

This view of basic research is much different than the view that it is primarily a type of self-realization or cultural activity. Where might Congressman Brown have gotten that idea? Perhaps he is taking at face value the claims of some parts of the physics community to have an exclusive franchise on fundamental research, a claim which all too often, whatever its merits, is allowed to blur into a claim to exclusive franchise on basic research. If so, we should now openly acknowledge that the phrase "basic research" has been stolen. And we should insist on getting it back.

We also need to add to the discussion, that despite this notion that the defining characteristic of basic research is new insight into nature, we should recognize that society values basic research primarily for the possibilities of control and exploitation that it sometimes affords. Thus, a company supports research, basic as well as implied, in the expectation of a competitive advantage for the corporation.

All other reasons are decidedly secondary. And the company uses this expectation of competitive advantage along with its insight into the future likely course of its products and markets to choose which areas of research are prudently viewed as good bets, as good investments-- in a word, as "strategic." That's what a company expects for research, including basic research.

The nation expects better health, secure defense, a better rising standard of living in an environment that has often been unnecessarily harmed more often. Research that can reasonably be expected to contribute to these goals, including understanding-driven basic research, is what I and the National Science Board's Special Commission on the Future of the NSF mean and meant by "strategic research."

There's a lot more that needs to be said about strategic research. And I will return to that in a minute. But I'd like to make some other remarks that perhaps put basic research further into context, as one of the several methodologies by which startling new discoveries may be made and new insights taken. The other major way by which new insights are occasionally obtained-- in fact, I would say 30-40% of the time-- is as an unexpected byproduct of applied research.

In the course of trying to make something work, it occasionally happens that questions are implicitly put to nature which call for unexpected and exciting insights. Two of the many possible examples are the discovery of radio emission from the heavens by Carl Jansen and the discovery of the 3-degree Kelvin cosmic background radiation by Penzias and Wilson. As I'm sure you know, both of these wonderful discoveries were made in the course of very applied and directed research.

A further remark that helps put the discussion of basic research into context is the following-- even in areas where new research can lead to new products and increased productivity to the achievement of society's goals, getting the R&D running, so to speak, is not more than 3% to 5% of the job.

There is at least 20 or 30 times the effort required in a whole complex of other activities-- development, manufacturing, market development, marketing, financing, advertising, pricing-- all of which have to be done competitively, that is quickly and well, if there is to be any lasting competitive advantage to the company or the nation which supported the research in the first place.

We have a failure that we share with the rest of our fellow human beings, which is that we tend to overstate the importance of what we do. There are many cases where the new insights from basic research are not one thousandth of the job.

Now, we have not only to understand the importance of these other activities, but also to give them a degree of respect, which is all too rare for us to do. What we do will, in the final analysis, not really help those whose tax dollars supports, unless these other tasks are done in a world-class way.

In addition to understanding and respect this, there may be other ways in which we scientists might help. And I will return to this subject in my third Compton lecture entitled What Is A Science PhD For. By the way, the observation that getting the R&D right is only a small part of the job is helpful in putting this question in front of us-- how much should be spent? In a word, how much research is too much?

Now, I had to think long and hard before I even put that question before you on the grounds that that might lead to my immediate expulsion.


A corporation should not over the long term invest more in R&D than it can successfully turn into added value for its customers and shareholders. Limitations in the ability of the firm in the non-R&D arena may limit what can be capitalized on from R&D, or it may be that the marketplace is such that it no longer conveys an advantage that it may once have done to be a leader in R&D.

Thus, a poor value for a company and I think for a country is to spend as much as you could possibly afford. Better to make sure-- or I won't say "better." I would say it is just as important to make sure that the other 95% of the job is in good hands and under control as it is to worry about, at this point, further increases in R&D.

Now, I'd like to come back to the notion of strategic research, which is in great need of clarification. It is not a category which is of primary interest to the duers. It is a category of interest to the supporters of research.

Recall that strategic research is research in scientific areas which are good bets to be helpful in reaching agreed upon goals, whether of a country or of a corporation. It's the task of those who support research with the idea of achieving goals to support the complete spectrum of research, from the most basic through engineering and applied research in those areas.

Perhaps a bit of history will help. Much of the university-based research establishment of which we are so proud and about which we are so worried was created and sustained over the last 40 years. I'm talking not about the health research establishment, but physical sciences. Was created as a strategic investment in superior military assistance. That's not true of all of it, but it's true of most of it. It was certainly true for the renaissance of modern optical physics sparked by the laser [INAUDIBLE].

As Professor Charles Terence of UC Berkeley has pointed out on many occasions, his early work on microwave spectroscopy was not part of a program of strategic research. It was part of what one might call discipline-only-driven basic research. But once the microwave laser had been invented and demonstrated, even though at that point quantum electronics was still really in its infancy, it was clear to the sponsors that they should increased their support for basic research in quantum electronics.

Because it had become a very good bet indeed that there would be applications of real value to the military. It had become strategic research almost from the word "go." The motivation of the researchers did not change a bit. They were still motivated by the desire to understand it. But what they did was in a different class as seen by the sponsors.

Similarly, Professor Nico Bloembergen, his pioneering work in nonlinear optics, with which I had the joy to be associated, was no less pioneering for having been part of a strategic research on the Joint Services' electronic program.

This an aside. It won't be in the published text. I'll never forget what are the annual reviews when the Joint Services electronic command came to Cambridge to review this. And Nico stood up and introduced the proceedings by saying, there are three types of intelligence-- human, animal, and military.


But those who supported him and others came to see that there was a high likelihood that many useful things would come out of this exciting and rapidly expanding field. They backed those hunches with support for strategic basic research. These examples ought to make clear there is no inherent conflict between strategic research and research done by individuals with a desire to understand nature.

This is an observation, by the way, which I direct as much at the funding agencies and Congress as at the scientific community. Moreover, since strategic research includes the entire spectrum from basic to applied, there should be no reason to talk as if the standards of peer review, or quality expected, or the participation of top researchers will somehow be less in strategic areas than in discipline-only-driven areas.

Now, to be candid, those of us who advocate the use of a category such as strategic research are well aware of the reluctance of many in the scientific community to embrace this notion today. We're also aware of the real difficulties to be faced if one takes this category seriously and tries to use it to shape debate and policy. I propose to deal with the fear first and the difficulty second.

So far as I can tell, the fears are of two kinds-- first, that the category of strategic research, including strategic basic research, will lead to disproportionate funding for such work to the detriment of what I've been calling discipline-only-driven research. Second, there is the fear that misunderstanding of the complex interplay between basic and applied research will lead in the strategic areas to the requirement that even the basic research be done according to the agency-driven timetables and milestones.

Those of you who followed the recent discussion on the Senate Appropriations Committee report on the NSF will appreciate that there is a real fear here. We must fight this notion on the grounds that it doesn't work, and it doesn't work because it doesn't reflect the way in which nature gives up her secrets.

As I said, the dangers are real. But it may help us as a nation to avoid them and still make proper use of the notion of strategic research if we reflect on some of the important differences between basic research in areas strategic for military systems and the support of basic research in areas thought to be strategic for civilian goals-- improved competitiveness or a better environment. Many of these issues were discussed at length in that earlier lecture that I mentioned.

The role of the government vis-a-vis research in support of military systems was and is very different than the possible roles of government as a supporter of research in aid of competitiveness or other civilian aims. The government is the customer for military systems, and as such can guide, and control, and support research by whether it's clearly-understood needs are being met or are likely being met.

But the government is not now and never will be the definitive customer for most of the civilian economy. This really changes matters. It is not expert on the results to be expected. Moreover, at least for 25 of those 50 golden past years, the military did not have to conduct the bulk of its support or its reasoning about research support under the public eye.

Parenthetically, the fact that the strategic motives of the military's supporters of basic research were not subject to constant public debate has had the unintended consequence that several generations of physical scientists were trained and have done much of their work without ever having had to think about why they were supported. That, of course, has changed dramatically in the last two or three years.

Another key point from that earlier lecture was to point out the deficiencies in US industrial competitiveness. It had almost nothing to do with lack of speed technology transfer from university labs or government laboratories. Lack of competitiveness stems from failures of investment, lack of attention to low-cost manufacturing, lack of attention to training-- to many other factors which are failures of corporate management, or failures of macroeconomics, or both. These will not be cured by a better technology transfer.

By the way the overwhelming-- and I say this as a long-time user of the product of the university. The overwhelmingly most effective mode of technology transfer from the universities to the civilian economy is via the trained students which you send out each year.

By the way, the lack of that mechanism is why I have almost no hope for real value in technology transfer from the national laboratories. Unless we turn it around with two years [INAUDIBLE], we are not going to get what they may have. That remark further drives the nail in the coffin of any hope I might have had ever to have a job in Washington.


So there are these important differences in how the government must approach strategic research for the civilian economy compared to what worked so well during the Cold War for strategic research in support of military systems. And part of the distress that we all feel now is due in part to the fact that the consequences of these differences have not really been digested in the way either that the scientific community approaches the new strategic research or in the way that the funding agencies and the economists support it.

The other major fear caused by the phrase "strategic research" is that it will lead to the withdrawal of funding from areas of research not described as "strategic." There is a real danger here as well, but it can be confronted and I believe managed, though the argument is subtle and will need a great deal of explanation, illustration, and repetition.

As I've already said, I believe it is possible to pick out areas of basic research which are a good bet in helping society to achieve its goals-- say, in environmental remediation, but in many other areas. But it does not follow that because we can pick areas which are good bets, we know what areas of science are not relevant

We may be able to conclude-- in fact, I have concluded-- that certain areas are very long shots indeed. But it is the height of arrogance and folly to make that decision about all of the so-called non-strategic areas. Were we to do so, we would be voluntarily and perversely cutting ourselves off from likely sources of assistance.

A further reason why it is prudent to keep many of the discipline-only-driven fields healthy and vigorous is that they contribute to the health of the whole, and in particular to the health of the so-called strategic areas, including the basic research in those areas. This is another consequence of the fact that nature is not in fact very well described by the categories of our conventional scientific disciplines.

However, this argument-- and I expect we will all want to make it. This argument would be much stronger if we scientists had culture which was more encouraging of individual movement among [INAUDIBLE] disciplines, a point I'll come back to in the third lecture. But the main reason for vigorous support of the non-strategic areas is that we cannot know what is not relevant.

Perhaps this point is not well enough understood by those who make public policy. And we all have to say it over and over again.

Having dealt I hope with the fears, I want to deal now with the other issues here with difficulties of the notion of strategic research. Since nature is inexhaustible, and the number of fields, and sub-fields, and specialization seems to be nearly limitless, how can we rationally decide how much money should be spent and on which fields?

Here the sociology and culture of science are of little help. And it is here we may have to make some changes in expectations. By definition, we cannot use the criterion of understood utility. We do not have a methodology by which the scientific community can come to anything like a consensus in recommendations to those who make public policy.

That inability, I fear, severely compromises our ability to say convincingly to those in public office that basic research is a necessary part for the achievement of our national goals, even in areas outside strategic areas, and that it is a necessity, not a luxury.

As you will be aware, some of our colleagues give a simple answer-- support all fields to the extent that their members will feel themselves well cared for. I hope you agree that this is not helpful.


It isn't now. It never was. Now, the National Academy of Sciences, the National Academy of Engineering, and the Institute of Medicine have recently issued a report from their Committee on Science, Engineering, and Public Policy which begins to deal with these difficulties. It is called "Science, Technology, and the Federal Government-- National Goals for a New Era."

And I commend that report too. By the way, Phil Sharp was a member of the committee that wrote that report. It recommends that it be a US national goal to be among the leaders in all major fields of science and a clearly outstanding leader in what I've been referring to as strategic areas. In their language, "we should be the clear world leaders in those areas of science which may reasonably be expected to contribute to economic growth, health care, national security, and environmental protection."

The report goes on to recommend a process of comparative international assessment to determine who is in the lead in all major fields, along with an analysis of what is required to achieve or maintain the lead in areas deemed crucial to national goals. The report also suggests that these comparative assessments would turn up fields of science in which we are over-investing from the point of view of achieving national goals.

Presumably, the results of the assessments would be of the following type-- to the supporters of research, one would say, for example, our performance in Field A is not of the first rank, but it needs to be to help achieve our national goals. So its funding must be sharply increased. Our performance in Field B is world class. But that is an area which is not among the areas in which we desire to be world leaders. We will be satisfied to be among the leaders. Therefore, increased funding is not required.

One assessment I don't think I will ever live to read is the assessment that says, we're as good or better in this field as we need to be, and any extra money, if there is some, should go to other fields.


However, such an assessment would be a rational outcome of this process in my view. However, I doubt the ability of the current culture of science to support such a rational undertaking, at least as long as it involves decisions to reduce funding in some areas.

But the scientific enterprise in the United States is so large in absolute terms and so successful in most of its parts that the inevitable result of most assessments will be that we are among the world leaders, and since we have the world's largest economy, it is appropriate that we be no longer world leaders in most areas. So we will have achieved part of these new national goals. Assessments will be required, however, to make sure we do not slip.

A crucial and I think what would be a convincing step for the scientific establishment to take, and which would help counteract the view that we are just one more special interest and one more perhaps not too adroit special pleaders, would be to find ways to agree on the fields which are strategic-- which I don't think should be so hard.

And then we will have to find ways to agree upon the discipline-only-driven fields which should be supported, keeping in mind that some fields will generate insight and train young scientists much more economically than other fields. To do this, we will have to give up the view held by many that basic research is all of a piece, that distinctions cannot and should not be drawn, and that one should spend whatever it takes to exploit all of the good scientific opportunities.

This is a serious and important agenda in public policy. And our mission and the goals, in the report that I've been discussing, in my view is any discussion yet of who in authority should call for these assessments and who will own the results in the sense of expecting to act on them. But I believe it's a worthy direction for the nation.

If pursued and if successful, it will allow us to answer with conviction, leadership in basic research is not a luxury. It is a potential comparative advantage for the country that, at the same time, does a world-class job at all the other processes by which new insight is turned into societal value.

Moreover, we would have a rational way of allocating support for basic research not only in strategic fields, but also in the other areas of discipline-driven science, whose health is important to the whole, and from which experience has taught us may come unexpected contributions to the national welfare. I hope you agree that these are goals worth pursuing. I thank you for your attention. And I will be glad to respond to comments and answer questions. Thank you.


AUDIENCE: Would you be willing to tell me about specific projects, for example, the [INAUDIBLE] project to build a very fuel-efficient automobile, or a project like the Genome Project and the Super Collider? [INAUDIBLE]?

ARMSTRONG: Could you all hear the question? [INAUDIBLE] questions, but serious questions and very much to the point. What is my view of the idea that the government should band together with the big three automakers and work on a much more fuel-efficient car, and then in the same question, my view of Super Collider and the Human Genome Project.

Well, I'm enough of a tabletop scientist or a small-science scientist to be skep-- and I'm enough of a skeptic about the government so that I believe that the government has to be very careful about intervening in a major way in the civilian economy, which your first example was.

The example of Semitech, which some of you will be familiar with, is my idea of how the government ought to do this. It will be hard to get away with two or three more times. But basically, what happened was the industry got together and said, we need to work together, first of all.

And we need government support in improving manufacturing and technology for semiconductors. We'll put up half the money. We would like you to put up half the money, [INAUDIBLE]. And by the way, we in private industry will entirely determine the research program.


I don't think that's going to characterize the fuel-efficient car. There are already timetables and milestones. So this is an example of what I meant by saying that since the government is not really the market-maker, it's difficult for it to know how to intervene.

In the matter of the Super Collider, first of all, this is not a field in which I'm an expert. And there are certainly many at MIT and probably someone in here who are.

I fear we have gotten ourselves into a situation where we're damned if we do and damned if we don't. To push it through at this point is going to cause very grave harm to the cause of science that I've been discussing here generally. For it to disappear in flames will have another sort of very bad effect, I think, on the scientific stand.

It was folly in my view to make this a national [INAUDIBLE]. What national interest is there in knowing what God made in the first picosecond of creation? I mean, there is no conceivable national interest, as opposed to international interest. It is one of these things which this managing approach, this managing complex with respect to science is a dreadful [INAUDIBLE]. And we're paying for it. And all of science is paying for it.

About the Human Genome, I just don't know. I don't know enough of that one. It's hard to believe that all of that knowledge won't be damn useful. But whether there are alternative ways to do that, I don't know.

AUDIENCE: In your description of the National Academy of Science's report, it sounded as if you're advocating the funding to islands of mediocrity.


AUDIENCE: It sounds as if you're advocating the funding to islands of mediocrity, and that the worse you make your area look, the more money you'll get. And which individuals are going to receive that money?

ARMSTRONG: Well, I thank you for that clarifying question.


I certainly didn't intend that. And I assure you that the Committee on Science, Engineering, and Public Policy did not intend that. You see, what's implicit in your-- well, let's put it this way. It is implicit that they say in areas that are important to the nation, if we are not doing well, we should. Now, I would not choose to characterize that the way you did.


And I would not suggest that it is wise to characterize that the way you did. First of all, I don't believe there are any areas that will be found to be strategic in the sense that I spent so much time talking about it in which we are mediocre. There isn't much in which we're mediocre.

But damn it, the public is paying for it. And if there are areas of science that can reasonably help us achieve our goals, and for whatever reason-- disinterest, disdain, or underinvestment-- we have not been working there, we should.

AUDIENCE: I'm sorry. But who will the money go to? If an area [INAUDIBLE].

ARMSTRONG: It ought to go to the best damn-- it ought to be allocated in the same way you allocate everything else.

AUDIENCE: And that will achieve an improvement, you think?

ARMSTRONG: Well, do you believe in peer review or not? Quite frankly, my observation-- and this is meant as a friendly comment-- is that if you establish the budget, they will come.


Not only they will come, but they good ones will come.

AUDIENCE: But relevant to this, John, there is, at least from my point of view, an observation that an increasing part of the decision process in the allocation of resources has a political or geographic component partly fueled by the manipulations that occur in Honduras that you're not allowed to build upon and partly because there's a very clear wish that excellence somehow be spread around like a uniform butter across the country.

And I guess the question is, is it appropriate, for example, for every established state university to be a center of excellence in five or six areas? Is this a direction that, in fact, is leading us opposite of what you just said to the very best?

ARMSTRONG: Well, that's a very worrisome issue, Steve. It's one that I dealt with at some length in that earlier paper called "Research and Competitiveness." That was one of the problems of the new rationale. What I said in the paper was, when it was a question of military systems of better cruise missiles, it was okay for most of the money to go to MIT, and Caltech, and Johns Hopkins.

But if it's economic competitiveness that is the main driver, then no region of the US will think it has less claim than any other. I think that is a very serious worry. And the fact that we're going to have declining amounts of money to spread around in the face of these forces makes it worse.

But quite frankly, that's a problem that we have to solve apart from this other issue of, are we working where we ought to be working. And it's awfully easy for us here at MIT or at Harvard to say that the money ought to go to the best places and that we ought to chop off the fingers of the congressmen who support scientific pork. It's another thing to carry that day.

And part of the problem, I think, is that because we do not yet deal adequately with some of these issues that I raised today, we are perceived as special pleaders, and therefore we're in the pot, we're all wrestling, and that's how the game is played.


ARMSTRONG: Professor Jackie wants to know how somebody is going to educate the public. Well, I'm tempted to say come to my third lecture--


--on, what is a science PhD for, because that is I think part of the answer. Many of you know that better than I do. What I've been trying to address here today was ambitious enough. I've been trying just to say, within the family, so to speak, one are the things that we ought to be sort of reassessing and thinking about-- what it will take to improve the scientific literacy in a dramatic way-- you know as well or better than I do-- in children.

AUDIENCE: Doctor, I like your analysis a lot. There's a lot of rationality to it. But there seems to be one gaping hole in the whole story. At least the one that I identified has to do with your comment that when it came to research that is strategically important for military defense, the government was a customer, and therefore understood the needs and could assess the strategicness [INAUDIBLE].

And I agreed with that. Then you turned around and said that the government is not good at the civilian areas of strategic importance because it cannot assess those as well because it is not a customer. Who is going to do that? How's that going to be accomplished if not by government?

ARMSTRONG: Yes. That's a very important point in this logical argument. And I was rushing towards the end. But I did in fact in my text have something to say for that. Remember I said that one of the omissions in the report from the National Academies was they did not say who would call for and own as well as who would do this process.

Now, it can be done. That is, the whole premise of industrial research is that it is possible to pick out a set of scientific areas in which there is a good bet that you will get a payoff. So when it comes to the set of high tech industries, or improve health care, or environmental remediation, it is not a major job to figure out what areas are reasonably included in strategic. By the way, it's most of science below [INAUDIBLE] for sure.


But the more detailed assessments, I think, are not hard to make. The real question is, who will call for it and who will use it.

AUDIENCE: Who will pay for it.


AUDIENCE: And who will pay for it.

ARMSTRONG: Well, the issue is, who will pay for the research part of it. I mean, the federal government now pays for a whole lot of research. It is reasonable to assume that they would continue to pay for research as it applies to national goals as well as it applies to the health of science generally. There is not a mechanism to cause this analysis to take place at the moment. That is a hole. It's a hole not only in my exposition, but it's a hole in fact. I couldn't cover it up.

AUDIENCE: You made a very provocative statement today [INAUDIBLE] National Labs directors about the role of universities versus the National Labs. And we have a tremendous influx of people. And I wondered how you feel currently about the balance of support for students in the sense that now many graduate students who are doing basic research have their support come in grants and contracts that are controlled by professors or other potential investigators versus fellowship programs?

Do you believe that the balance is right? It seems to me that the balance is too much toward grants and contracts. And if we placed the money in the hands of students, it would distribute the resources geographically [INAUDIBLE] it does in fact bring a benefit to the whole nation?

ARMSTRONG: Mark, I think that's a tremendous point.

I think it is not only right, but I think it's-- how shall I say this? As a guest in [INAUDIBLE] and it's a particularly courageous point to make at MIT, where within a microsecond after enrolling as a graduate student, one is captured by a research program. But I think that is an excellent point.

As the president of the young investigative programs do, as I think the National Science Foundation fellowships used to do, if they are in the hands of the students and are not tied so closely in most cases to research careers, I think that would have a very celebratory effect.

One of the things that is missing in all of this-- to give you just a foretaste of the third lecture-- the interaction during graduate years between students in the university and their professors in the university, and the places where science is done in the outside world-- that interaction is in need of a great deal of strengthening. And having more money as scholarships as opposed to grants might facilitate new mechanisms to do that as well.

MILLIARD: Final question. There.

AUDIENCE: If high-energy physics is not a national goal, what makes it an international goal?


ARMSTRONG: Well, it seems to me that the-- let's put it this way. It really is interesting. It really is fundamental. It doesn't exhaust basic research, but it certainly is fundamental.

I think the probabilities of any sort of effective spin off at these extreme energies are remote. But if shared widely enough, I think it is a reasonable goal for some of our research. As I say, there's no reason to make it a matter of national pride or urgency. But I don't think that's equivalent to saying that we should turn away from it. We've just gotten ourselves into an untenable position. But I don't think we need to throw the baby out with the bath.

MILLIARD: It's time for the last question [INAUDIBLE].



MILLIARD: One fundamental concern I have in this [INAUDIBLE] into discipline-oriented research [INAUDIBLE] that, for example, I'd say 25 years ago, molecular biology would not have been considered strategic research. Nevertheless, one can see the impact it's had on the Massachusetts economy.

And I'm afraid that a more cynical way to define strategic research would be research that corresponds to [INAUDIBLE] improvements on existing technologies. And I think that's not the best game plan in the long run, and that in fact the probability of generating breakthroughs in technology are actually more likely to come from relatively undirected basic research. And I'm very worried when you use the word "strategic--" and I should say, as an aside, it happens that my own personal research is officially classified by the National Science Foundation as "strategic."


ARMSTRONG: I'd never doubt it, I'll tell you that.

MILLIARD: But I'm very worried when we use these kinds of words that we'll end up with all of the funding going into strategic research, and the kind of research that's really going to produce a breakthrough in the economy getting very little of it's share of the funding.

ARMSTRONG: Well, I understand the issue. And I dealt with it in my remarks. And I guess what you're saying is that you don't believe in our ability to successfully manage that distinction. And you may be right. But one of the points of my lectures was that if we have no conviction about this or don't understand it, it's hopeless for the world at large.

So I don't think we have a choice. And to say, well, there's this danger-- there is this danger-- and not say, by god, I can do just as basic research in a strategic area and somewhere else, and, by god, the health of the whole depends for the reasons that you know and that I said on the health of these other areas-- if we can't speak with enthusiasm and conviction about this, we're in very serious long-term trouble. So I would say we've got to all whistle in the dark a little bit more on this subject and get our convictions up.

MILLIARD: That was the last question.

ARMSTRONG: Thank you very much.