You have asked me to address the dangerous spillover from nuclear fuel plants that the President spoke about in his February 11 speech on proliferation.

To me the message is that however much the world needs nuclear power, it has an even more immediate and overriding security interest in stemming the spread of nuclear bombs. To limit the risks of their use we have to curtail worrisome nuclear activities. We have to do this intelligently and in a way that gains broad support around the world. Technology of Most Immediate Concern Today Is Uranium Enrichment

For many years our main concern in this area was the spread of plants that extract plutonium from used fuel. Once extracted, the plutonium is accessible for bomb use if that is what its owners intend. This is still a concern (one has only to mention the 8,000 missing North Korean fuel rods).

But the technology of most immediate concern today is uranium enrichment by centrifuge. Enrichment plants are needed to prepare commercial reactor fuel. But they can also enrich to much higher concentrations usable for bombs. A centrifuge plant that is small in commercial terms can produce a lot of bomb material, and it is easy to hide. A commercial plant sized to supply the fuel needs of one standard reactor could be reconfigured quickly to produce enough nuclear explosives for 20 bombs per year.

The depressing chronology of how centrifuge technology spread over the past decades from Germany and the Netherlands to Pakistan and from there to North Korea and Iran and Libya-and we can be sure this is only part of the story-provides us a list of what not to do if you care about stopping illicit access to bomb material.

Lax security at Urenco, the German-Dutch-British enrichment enterprise, allowed Pakistan to steal plans and contractor lists. Weak European export controls made it easy for unscrupulous contractors to supply the Pakistanis.

We did our share to enable Pakistan's bomb-making by looking the other way to encourage their help in the Afghan war against the Soviets. (A senior Pakistani nuclear manager told me they were pleased but astonished at the extent to which we did this.) Our intelligence agencies were slow to pick up what we now know-that the Pakistanis then proceeded to spread the technology.

The lessons for dealing with the clandestine trade are: tighter security, stronger export controls, no relaxation of U.S. anti-bomb priority, and sharper intelligence. Steps for Fixing the Nonproliferation Treaty

But closing the back door to the bomb, so to speak, is not nearly enough if we leave the front door wide open. The Nonproliferation Treaty does not specifically proscribe technologies or materials that bring a country dangerously close to bomb-making. It is this ambiguity that Iran exploits when it insists it has the right to enrich uranium and extract plutonium.

In the past we tried to plug the holes in the NPT system with ad hoc measures, but the system is getting too stressed for that. If we are going to use nuclear power, let alone expand its use, we need a new approach. As the President said on February 11, we can't let countries violate the purpose of the Treaty "under the cover of civilian nuclear programs." The Director-General of the International Atomic Energy Agency, Dr. Mohamed ElBaradei, says the same thing.

We need to:

This is a tough program to sell and any number of policy experts will tell you why it can't be done. But if that is so we had better prepare to live in a nuclear jungle. This much I know: nothing will be done to tighten the rules unless the United States takes the lead. Key to Gaining Broad Support for Effective Action Is a Common Standard for All

At the same time, we cannot realistically hope to get very far by ourselves. I believe the key to broad support for effective action is agreement on common standards, ones we ourselves accept.

This is the element I find missing in the president's February 11 speech. "Do as we say, not as we do," is not going to work.

And we can't have a permissive set of rules for governments we like and a different one for those we don't. How, for example, can we persuade that it is acceptable for Pakistan to enrich uranium but not acceptable for Iran? What do we say to Brazil, which has a substantial centrifuge plant underway? We can't convince the world by treating each country as a special case.

For a comparison in approach I'd like to take you back to President Gerald Ford's 1976 watershed statement on nuclear policy. Back then President Ford asked others not to extract and use plutonium at least until we had the international protection to make it safe to do so. The thing I want to underline is that he decided the United States would itself act in a way that was consistent with what we asked of others. "We must be sure," he said, "that all nations recognize that the U.S. believes that nonproliferation objectives must take precedence over economic and energy benefits if a choice must be made."

As it turns out plutonium fuel is highly uneconomic and nuclear power doesn't need it, so it has become easier to act consistently with what we ask of others. Despite this, the Department of Energy, with the support of the administration's energy plan, has been trying to loosen the restrictions on the technology to ease the way for plutonium-related reactor programs in the United States. I believe the US-Russian program to use surplus plutonium to fuel civilian reactors falls in the same category. It is justified on the basis of nonproliferation, but it is in my view principally a way of getting the plutonium camel's nose in the tent. If we continue with these DOE programs we will encourage plutonium use world-wide, including in some chancy places. Proposal for a Common Standard for Enrichment

The enrichment problem is tougher because modern reactors cannot do without enrichment. (There is, however, no need for bomb-grade enriched fuel that is now still in use at some research reactors.) If centrifuge enrichment plants become widespread we will find it very difficult to ensure that they are all restricted to low enriched output, or to keep track of their output. It will become especially difficult to find clandestine plants because their indications would be masked by those of commercial plants. To add to our worries, if a would-be bomb-maker started with low enriched uranium instead of natural uranium, his work to reach the highly enriched bomb grade stuff would be reduced by a factor of five. The only sensible answer is to restrict the number of centrifuge operations to a few large ones. But how can we persuade countries to go along with this?

I don't hold much store in the various proposals for multinational fuel centers that have occupied academic seminars for years but lead nowhere. I have to say I am also skeptical about the President's fuel guarantee proposal. I think it would end up as an ever-growing carrot. The best guarantee is a market with several competing commercial suppliers.

A way of squaring this desired result (only a few large centrifuge operations) with the application of a common security standard is to make centrifuge enrichment plant owners pay the large safeguards cost they impose. A hefty minimum would make small plants uneconomic to operate. Encumbering small centrifuge plants in this way would be fair because the international security cost they impose is way out of proportion to their size. Need for Continued Close International Inspection

It would be nice to end by saying that if we limited commercial enrichment and reprocessing we wouldn't have to worry about the reactors themselves. Unfortunately, mainly because of the concern about small clandestine reprocessing plants, reactors and their fuel will always need close IAEA oversight, especially those located in iffy countries. Restricting fuel cycle plants will however make the inspectors' safeguarding job manageable.

There are no easy answers. Above all, we have to keep our security priorities straight, and to act in such a way that everyone understands them.