APS News

The Back Page

Balancing National Security Concerns with Scientific Openness

By Michael May

Michael May
Michael May

There has been a recent flurry of committee reports and public statements - the Cox Committee Report, the Rudman Report, and the Zachariah report, among others - pointing to alleged security deficiencies at some US defense and space contractors and at the US nuclear weapons laboratories. Specifically, it is alleged that a cleared Los Alamos employee, a US citizen of Chinese origin, has given secrets to China; that China is using or trying to use all US residents of Chinese origin to steal secrets; that US firms such as Loral and Hughes have passed secret or sensitive missile guidance technologies to China in the course of using Chinese missile launch facilities; and that the Department of Energy (DOE) has for many years stonewalled efforts by its own leadership and by others to remedy faulty security procedures.

The accusations against individuals have been contested and are not publicly documented. Several individuals involved in the investigations have testified that racial prejudice played a role in the investigations. Bureaucratic and other politics clearly played a role as well. Many of the technical and historical specifics that have been released are wrong.

Any enquiry into the subject of scientific openness and national security, but any strategy must take the present political environment into account. However, it must also take into account our earlier long-term experience with balancing national security and scientific openness, our goals in carrying out this balancing, and our successes and failures in the attempt. This earlier experience tells us a different story and counsels a different strategy from the story and strategy we might derive from the headlines.

The first thing history tells us is that the US has been extraordinarily successful in applying science and engineering to defense, space and other military systems. The missteps and false starts of course made headlines, but the relentless, open criticism of mistakes and failures is part of the reason for the US success. Today our defense technology is the best in the world.

The second thing history tells us is that the US has run by far the most open defense R&D establishment in the world. The openness is directly connected with our success. Immigrants, refugees, and foreign-born citizens have been welcomed into the most secret circles of classified R&D, and they have made pivotal contributions - specifically in the fields of nuclear weapons and missile technologies, where openness is currently under attack.

The US has repeatedly told the world what it was doing and why. We have debated the most delicate of defense matters in the most public of manner. We have taken the lead in sponsoring international meetings, arms control measures and other initiatives which require international technical interaction. We have sent our cleared scientists and engineers freely all over the world. No one else does quite the same thing.

It might seem to some as if the US were allowing others to catch up. But, if so, why haven't they caught up? The US has done this from the early years following World War II. Not only have the others not caught up, they have fallen further and further behind, not in civilian technologies, but precisely in the defense technologies we are talking about today.

Why has this happened? The very large US investment in science and technology is certainly a principal reason, but it can't be the whole reason. Why have others not taken advantage of our investments to take a shortcut and get ahead? Why did the Soviet Union, with its large spying apparatus, its extensive secrecy, and its very large investments in defense, never come close?

Visiting the Soviet Union in the old days showed what was happening. They had barriers between laboratories and design bureaus, barriers between ministries, between cleared and uncleared scientists. Heaven forbid that Soviet cleared scientists should talk with foreigners. As a result, they wound up not taking advantage of their best ideas. In many cases, we did. Many of these barriers exist to a lesser extent in other countries.

Large investments were essential to the US lead but they were not enough. Some of our large investments were huge mistakes. Open criticism and input from people outside the cleared communities have been essential to put us on the right track. These interactions are even more badly needed today when so much is happening outside the defense science and engineering communities. They are essential in the weapons laboratories, to support the changing science and technology base of the classified projects, to carry out such international tasks as fissile material control and arms control verification, and to retain and hire good scientists and engineers. In the missile business, where American firms must both remain in the lead in their defense tasks, and also field competitive commercial systems such as satellite packages, interaction is even more essential.

What is the right strategy for scientific and engineering organizations to pursue in order to help in the current debate? First, as many people have noted, the alleged breaches of security are not connected with interactions of cleared personnel with the general global science and engineering community. Neither do they involve the spread of unclassified information. The relevant allegations involve cleared US citizens and matters like the design of nuclear weapons, which, by any standard, should be kept secret. Let's put our efforts where the problems have been.

Second, to protect essential secrets and at the same time interact with both foreigners and the American scientific and engineering communities at large, defense personnel must have clear and sensible guidance. That's the way it's been in the past. A clear line used to be drawn around the information that could and should be kept secret, with some margin so that this information would not be partially given away by its boundary.

This clear distinction is being lost. A fog has descended on what used to be a useful workable boundary. New categories are being defined or redefined which blur that boundary and will make it difficult or impossible to carry out technical interactions. Material which has long been in textbooks around the world, such as the fusion cross-section of tritium1, or basic missile guidance principles, may now require a permit for discussion, with heavy penalties to be imposed if unclassified information, generally available but deemed sensitive, is inadvertently released.

This trend must be turned around. It leads to a situation that will not be workable for either technical people or for professional security people. It is the contrary of what the US has done in the past, and much closer to what the Soviet Union and other police states have done. It will hurt not only the scientific and engineering communities, but also the students and the people at large. It won't hurt the spies.

Third, spies are caught through good line leadership and good security work. Security professionals have repeatedly caught spies. The last bunch came to public attention about ten years ago. None of the constraints now being discussed would have helped catch them. But those constraints will hurt the work of security people by diffusing their responsibilities and by damaging their relationship with the technical community, a relationship which up to now has been good, and which is essential if the counter-intelligence people are to do their work.

The other part of good security is a responsible and empowered line leadership, in Washington and in the field. Laboratory directors, company presidents and agency heads should be responsible and empowered to maintain security, just as they should be responsible to maintain operational safety, meet environmental standards, follow the laws and national policies in personnel practices, and do all the other things through which our technical enterprises discharge their responsibilities to the public at large. Of course, there must be effective monitoring and reporting outside line channels. But czars are useless.

I would suggest that a joint group of security professionals and people from the technical communities concerned help draft any needed new security procedures for interactions with the open communities and with foreigners; and that people who have a record of success in keeping the US ahead in the defense technology areas help draft any change in administrative and reporting structure. Clearly the political authorities will have the last word. But any new initiative, such as the new agency now under consideration, can be done in such a way as to help or hinder the basic objective of keeping the US ahead. The devil is in the details. This should not be just the result of a political compromise.

Regarding export controls, the US no longer has an overall structure and philosophy for such detailed practices, which are often conflicting and self-defeating. It is no longer possible to have an export control policy as clear-cut as our Cold War policy. The US must cooperate, learn from and help countries that are partners in trade, in international initiatives of the first importance, and in many technically oriented matters, but which may also be now and again political, economic and military rivals. Nevertheless, some guiding principles of an exports control policy can be laid down. Detailed laws and regulations can then follow those principles, with frequent scheduled revisions to take care of changes.

Those guiding principles would include protecting what is key to our current military edge, which is first of all the people and teams involved, and also includes specific system design and test information. The people come first, both for technical achievement and for security. Another guiding principle is that the US, because it invests more, usually profits more from common knowledge, so it's to our advantage to increase the pool of common knowledge. Clear lines around what is essential to protect should be drawn, but outside those lines, we must allow for this common knowledge to grow.

Our system is to run wide open. We let all students come and learn, not just science and engineering, but a lot about how the US works and what the world outside their countries is like. We let our cleared scientists and engineers participate abroad, talk, listen, buy and sell. It's a great system for all, but we derive the greatest benefit from it. It's worked both to inform our people and press, and to put our defense technology ahead of everybody else's. Other nations have used the system of suspicion, widespread secrecy, and "gotchas." Look where they are today. Let's not let inevitable rivalries cause us to trade our system for theirs.

Michael May is a professor of research in the school of Engineering at Stanford University. This article was adapted from a talk delivered to the American Association for the Advancement of Science on August 31, 1999, in Washington, DC.

APS encourages the redistribution of the materials included in this newspaper provided that attribution to the source is noted and the materials are not truncated or changed.

Editor: Barrett H. Ripin
Associate Editor: Jennifer Ouellette