-in cooperation with the Energy Research and Development Administration, research on the safeguarding of isotope enrichment plants; and

-in cooperation with Canada, the United Kingdom, the Federal Republic of Germany, and the Netherlands, the testing of new instruments to monitor nuclear facilities.

With respect to the timeliness of safeguards the prospects for improvement critically depend on the directions in which civil nuclear technology evolves. (This problem of timeliness is discussed below.)

IAEA safeguards apply only to "declared" facilities. With respect to unsafeguarded or undeclared facilities we must depend on a strong national intelligence capability. Quite apart from the problem of undeclared facilities, the United States must rely on its intelligence resources to assess the proliferation of nuclear weapons capabilities. Indeed, for the whole range of U.S. policies against nuclear proliferation, a strong national intelligence capability is vital.

The Nonproliferation Treaty (NPT)

This treaty was co-sponsored by the United States, the United Kingdom, and the Soviet Union for the purpose of controlling the spread of nuclear weapons while facilitating peaceful uses of nuclear energy. Negotiated primarily in the Geneva Disarmament Conference during the 1960's, the treaty had 98 parties and 12 other signatories by June 1976.

Under the treaty, nuclear weapon states agree not to transfer nuclear explosives or control over them to nations which do not have them; nonnuclear weapon states agree not to acquire them and agree to place all of their peaceful nuclear activities under safeguard agreements and verification procedures of the International Atomic Energy Agency; all parties agree to require such safeguards on nuclear exports to any nonnuclear weapon state. The treaty also provides for cooperation in the peaceful uses of nuclear energy and for negotiations in good faith toward nuclear arms control and disarmament. There has been no known violation of the treaty.

Some important potential nuclear weapon states have not signed or ratified the treaty. But the majority of states with the potential to develop nuclear weapons, or to export the requisite materials, equipment and technology, have a strong commitment to the goal of nonproliferation. Most of them are parties or at least signatories to the NPT. Moreover, with a few important exceptions, all existing nuclear facili

ties in the world's non-nuclear weapon states are now under international safeguards.

While the treaty itself does not impose sanctions against violators, the statute of the IAEA makes provision for sanctions against violations of the safeguards required by the treaty. And while the treaty permits a party to withdraw on 90 days' notice, if it decides that extraordinary events related to the NPT's subject matter have jeopardized its supreme interests, the strong mutual interests of the parties in preventing further proliferation would increase the political costs of exercising this right of withdrawal.

The treaty makes a continuing contribution to nonproliferation efforts. Perhaps its principal benefit is that it supports and links together the commitments of many countries not to develop nuclear weapons. It formalizes understandings among neighboring countries not to start nuclear arms competition with each other, and it codifies internal political attitudes against initiating a nuclear weapons program.

The NPT Review Conference held in May 1975 produced a number of significant recommendations. It expressed strong support for IAEA safeguards and called for greater efforts to make them universal and effective. There was general agreement that safeguards do not hamper peaceful nuclear activities. The final declaration of the conference, accepted by consensus, urged all states with peaceful nuclear activities to maintain effective accounting and control systems, noting the readiness of the International Atomic Energy Agency to assist them and the importance of such systems to effective IAEA monitoring. It recommended that safeguards be of adequate duration (which had not always been the case in the past), and that they preclude the diversion of safeguarded material to any nuclear explosive device, regardless of its stated purpose. It recommended that support be given to improving safeguards techniques, instrumentation, data handling, and implementation.

The conference urged common export requirements designed to extend safeguards, in all available ways, to all peaceful nuclear activities in importing states not party to the treaty, and urged all suppliers and recipients to accept these requirements.

The conference recommended better physical protection of nuclear materials, giving weight to NPT adherence in decisions on nuclear assistance and credit arrangements, and also recommended the study of the potential advantages of multinational fuel cycle centers.

The conference emphasized the need for further study of "peaceful" nuclear explosions (PNEs) by the IAEA (which subsequently established an international advisory group on this subject) and for considering the arms control implications of PNES by the Conference of the Committee on Disarmament (which subsequently held a meeting of experts on this aspect of the subject).

Export Controls

Perhaps the most striking achievement of the review conference recommendations was the extent to which they laid the groundwork for the subsequent understanding among major nuclear suppliers on export policies. Recent years have seen the emergence of several nuclear exporting nations. Competition among suppliers has sometimes stimulated sales which enhance the weapons potential of nonweapon countries. The review conference exerted a strong influence on supplier nations to recognize their mutual interest in controlling exports.

In recent months the United States has engaged in bilateral and multilateral discussions with nuclear exporters to develop common rules on export controls and safeguards. As a result, the United States, together with other exporters, has decided to apply certain principles to future nuclear exports. These are designed to inhibit the spread of nuclear weapons while permitting nuclear exports to help meet the world's energy demand.

Under these principles exporters must: (1) apply international (IAEA) safeguards to all nuclear exports; (2) prohibit the use of these exports to make nuclear explosives for any purpose-whether called "peaceful" or not; (3) insist on adequate physical security for nuclear facilities and materials to prevent theft and sabotage; and (4) demand the same conditions on any retransfer of these materials or types of equipment to third countries. The suppliers lent encouragement to the development of multinational facilities and agreed to special conditions governing sensitive goods and technology.

Beyond accepting these minimum standards, the United States intends to exercise more stringent self-restraint in the supply of sensitive exportsthose which involve fuel enrichment, spent fuel reprocessing, and heavy water-when they could add to the risk of proliferation.

Physical Security

With the objective of preventing the theft of nuclear materials, a group of experts from 11 countries

and the Euratom atomic energy community met in 1975 to revise the IAEA publication, "Recommendations on the Physical Security of Nuclear Materials." The result of their deliberations was the issuance of IAEA Information Circular 225, "The Physical Protection of Nuclear Material." The revised document is more comprehensive and explicit than its prede

cessor.

The United States has begun drafting an international convention on physical security. It takes particular heed of the necessity for international understandings about arrangements for the security of nuclear materials during international transit.

Multinational Nuclear Centers

Under the multinational nuclear fuel cycle center (MNC) concept, a number of nations would share facilities in a given location. These facilities might be used for one or more purposes: management of waste, spent fuel storage, fuel fabrication or storage and-more uncertainly-for various phases of the provision or storage of readily fissionable material or fuel elements involving them, such as plutonium reprocessing or the fabrication of fuels using plutonium. This might allow the economic benefits of large scale operations; improve security against theft and terrorism by reducing vulnerable transportation links; and most importantly, reduce suspicion and fears of secret nuclear weapons competition while providing a more effective safeguards system.

The idea for such centers was advanced at the NPT Review Conference and by Secretary of State Kissinger before the U.N. General Assembly in September 1975. The IAEA is now studying the concept, ACDA and others are providing expert consultants, and ACDA is also carrying out studies of its own on a broad range of related questions.

The MNC idea must be explored-together with other alternatives-thoroughly and cautiously since it involves many new economic, technological, and political questions. But this exploration must move ahead without delay if it is to provide a useful alternative to national programs.

Will nuclear weapons continue to proliferate? The threat is more serious than ever before, and our efforts to stop proliferation are stronger than ever before. No one knows what the outcome will be, but we are improving our understanding of the key factors and how we can influence them.

During the past year ACDA has funded private studies of long-range trends in nuclear proliferation. These studies are examining ways in which further

proliferation might occur, policies that can reduce the likelihood that it will occur, problems that would emerge with further proliferation, and policies for dealing with those problems.

An examination of ways in which proliferation might occur suggests two major uncertainties that affect the future of nuclear proliferation:

-the extent to which civilian nuclear technology evolves toward practices that maintain substantial lead-times between potential safeguards violations and the manufacture of nuclear explosives.

-the extent to which the world community acts decisively at "turning points" in the proliferation process.

By making prudent choices we may be able to influence both of these uncertainties. We shall now examine each of them.

The Problem of Short Lead-Times

The diversion of nuclear fuel or nuclear technology from peaceful purposes to the manufacture of weapons is contrary to most of the agreements covering the transfer of such materials. Such agreements, however, do not plainly restrict a nation from building a stockpile of nuclear weapons material that is subject to IAEA safeguards. Only the spiritrather than the letter-of these agreements would prevent national efforts, not involving this material, to design weapons or to develop nonnuclear components of nuclear weapons.

A termination or breach of IAEA safeguards would still be necessary to fabricate nuclear materials into nuclear explosives. Sufficient warning of such an event is vitally necessary to allow effective response. If the critical time-the period required actually to produce a nuclear weapon-is short, while warning and response required a long period, a nonnuclear weapon state might produce at least a small number of nuclear weapons well before the warning was sounded.

The principal facilities of current civilian nuclear technology-the light water and heavy water power reactors-by themselves do not bring a nation closer than many months or years to nuclear explosives. However, other related facilities (such as reprocessing plants) and some potential new developments could make weapons-usable material much more readily available. This could shrink the lead-time to weeks or even days, if all the other elements required for manufacture of nuclear explosives were in place.

The present international safeguards system can give timely detection and warning when many months or years are required to produce nuclear explosives. But it is far less certain that safeguards can provide timely warning of diversion attempts when much shorter lead-times to weapons production are involved.

The risk of detection alone is not a sufficient deterrent to proliferation. Timely detection-as cited in the Presidential Report of 1975 on international safeguards and earlier by the IAEA-is the essence of adequate warning to allow an effective response by the international community.

The timeliness of safeguards must be improved, but these efforts could be frustrated by new technologies. The prospects for technologies that will shorten lead-times depend on technical, economic, and public policy considerations. The influence of many of these factors differs between advanced industrial states and underdeveloped countries. Since the risks of proliferation grow with the shortening of lead-times, we shall examine the prospects for three of the more important technologies: plutonium recycle, breeder reactors, and peaceful nuclear explosions. The first two of these technologies could reduce the lead-time to nuclear explosives to days or weeks; the last reduces the lead-time to zero.

Plutonium Recycle

Many specialists assert that plutonium and unburned uranium should be separated from spent reactor fuel and reused (Figure 2). It is true that these fissile materials could replace some of the new uranium fuel needed, but reprocessing could at most supply only about one-third of the fuel required and far less in a growing nuclear power system.1

There are various uncertainties in the economic case for plutonium recycle compared with power generation without recycling. Although the cost of new uranium and enrichment for reactor fuel has gone up, the cost of separating plutonium from spent fuel has gone up very much more-tenfold in only 10 years. Relative costs are likely to continue to fluctuate, but it now costs more to separate plutonium for reuse than the plutonium is worth as a fuel.

1 A recent projection by researchers at the Nuclear Energy Agency of the Organization for Economic Cooperation and Development indicates that plutonium recycle could delay a given level of uranium consumption only by about one year in the late years of this century and would reduce annual uranium consumption by less than 10 percent. These figures agree closely with earlier projections by the U.S. Atomic Energy Commission.