New Trends in Safeguards

In 1980, I presented a paper entitled \"Prospective Trends in Safeguards\" at an ANS Conference in Mexico City. My inclination, having plagiarized its title, was to borrow its text as well. On re-reading the 1980 paper, I decided this was impractical, not because it would be remembered, but because much has happened in international safeguards since then. All of us are familiar with two events, both unprecedented in their way, that highlighted in different ways the role of safeguards in the nonproliferation regime. The first of these was the 1981 bombing of the Osirak research reactor in Iraq, an event which—mistakenly in my view--was characterized even by some IAEA officials at the time as an attack on Agency safeguards. The second was the Agency's own determination, for the first time in safeguards history, that it was no longer in a position to verify that no diversion had occurred at certain facilities subject to Agency safeguards. Both of these events had important implications, not the least of which was a predictable resurgence in interest in safeguards on the part of Congress, the media, critics and the safeguards community itself. While some would differ with this assessment, I believe that, on balance, the debate engendered by the Osirak attack was a healthy one for international safeguards. In the Pakistan case, while noone would contend that a finding of non-verification is good news, the demonstration of the Agency's political capacity to reach such a conclusion and submit it to its governing Board was a major step in the evolution of international safeguards. Important as these events were, the major changes since 1980 have come about not in the form of dramatic occurrences but in a series of small steps which, in the aggregate, add up to substantial progress. They provide us with an international safeguards system which works significantly better and, not unimportantly, is understood significantly better by its practitioners and others than was the system of 1980. I want to go to some pains to make this point in order to avoid any implication that the comments that follow are intended to express dissatisfaction, much less lack of confidence, in Agency safeguards. The lively interest in the safeguards community today to seek still further improvements in the effectiveness and efficiency of safeguards is in itself a very positive factor. In my talk today, I will try to identify several concrete areas where I believe further evolutionary progress is possible, and, in many cases, already underway. Before doing so, I want to note that this analysis owes a great deal to work currently underway on the part of IEAL for the Arms Control and Disarmament Agency, under a contract to explore possible safeguards improvements for the years ahead. However, the views expressed in this paper are exclusively my