The Soviets have recently made a new strategic arms control proposal at Geneva, and they have also continued to press the United States to agree to a Comprehensive Test Ban -- that is, a prohibition on any underground tests of nuclear weapons. The administration has stated that it is examining the arms control proposal but has rejected the CTB.

Some who advocate a CTB maintain that a ban on testing would generally lower the reliability of nuclear weapons and thus advance the cause of peace by making all such weapons less usable. Other CTB advocates believe the contrary -- that testing is not really needed to ensure weapon effectiveness. Still others see a CTB as important chiefly as a symbolic step toward the eventual abolition of all nuclear weapons.

Not only do these arguments in part contradict each other, but none holds much water. The hard truth is that for the foreseeable future the stability of the strategic balance and the security of this nation and the West will depend upon the deterrence provided by nuclear weapons. Neither good intentions nor a single technical fix (whether it be a CTB or, for that matter, deployment of space-based ballistic missile defense) will lead to nuclear weapons' becoming obsolete within any relevant time frame.

That day, if it ever comes, can only be brought about by Western strength and by the Soviets' gradual acceptance that the perpetual expansion of their power is a hopeless goal. Today, however, nuclear weapons reflect the political chasm between freedom and totalitarianism. As long as this political competition exists, a strong military posture will be required to deter Soviet aggression and expansionism. In particular, although we may be able to reduce our degree of reliance on nuclear weapons and improve their safety -- and although we may hope for reasonable arms control and subdued rhetoric -- such weapons will be required for an indefinite period to preserve the peace. Our allies appreciate the contribution Western nuclear weapons make to peace and stability today, perhaps better than we do.

In light of this, it would be a serious mistake for the United States to abandon nuclear testing now. First, limited testing permits the development of weapons that are far less sensitive to the explosion of their conventional explosive components as a result of a terrorist incident or an accident. Even if the actual conventional detonation of a nuclear weapon is highly unlikely in such cases, the resultant scattering of nuclear material would be devastating.

Second, testing is required to replace high-yield weapons with new lower-yield ones -- a change made possible by the more accurate delivery systems of today. These sorts of modernization of nuclear weapons contribute to strategic stability.

Finally, experts from the U.S. nuclear weapons laboratories largely agree that, over time, some testing is required to ensure reliability of the nuclear stockpile. Lack of confidence in the stockpile would impair our ability to rely on nuclear weapons for deterrence. And reduced confidence in reliability under a CTB might well be of greater disadvantage to the United States than to the Soviets. With our greater reliance on quality than on quantity, our confidence in our weapons might well deteriorate much faster than the Soviets' confidence in theirs.

Moreover, there is substantial debate about how to verify a CTB treaty. A threshold about one-tenth of the current 150-kiloton limit could probably be verified adequately with seismic stations outside the adversary's territory. An even lower threshold could be verified with cooperative measures that permitted each nation to have seismic stations in the other. But some cheating is possible at the level of very small nuclear detonations. Moreover, at these very low levels the borderline between nuclear tests and directed energy experiments becomes difficult to define. The inevitable political debate over verification of a CTB would add little to our understanding or to strategic stability.

The argument is often advanced that a CTB will slow the spread of nuclear weapons to other nations. It is true that American and Soviet nuclear testing provides for nonnuclear states a convenient political rationalization for pursuing nuclear weapons development. But the reasons that lead a state to acquire nuclear weapons are fundamentally grounded in that state's perception of its own security interests, not in the behavior of the United States or the Soviet Union. South Africa today, for example, is unlikely to avoid nuclear weapons development because of a U.S.-Soviet CTB. Adherence by the superpowers to a CTB will change the rhetoric of some nations about nonproliferation but will have at best a marginal impact on the reality of what they do.

A reasonable policy for this nation to follow would be, first, to ratify the Threshold Test Ban Treaty signed by the United States and the Soviet Union in 1974. Following that, the United States and the Soviets could enter into negotiations to lower this limit to a level consistent with each country's technical requirements and its ability to verify, with high confidence, compliance with the lowered threshold. This would exert a long-term pressure toward lower yields and less reliance on nuclear weapons without the attendant problems described above. Finally, both the Soviets and the United States should renounce the use of nuclear explosives for so-called peaceful uses.

This set of proposals will fully please neither side in our current domestic debate nor does it comport with the current Soviet proposal. But it offers a reasonable approach toward dealing with this difficult issue while we devote principal attention to the main show -- maintaining a deterrent and pursuing stabilizing strategic arms limitation agreements.

Brent Scowcroft is chairman and John Deutch and R. James Woolsey are members of the President's Commission on Strategic Forces.