The purpose of a major strategic arms agreement should not be to implement a slogan, to flatter the ego of an outgoing administration or to require us to bet our whole strategic future on a gamble, such as early SDI deployment. The purpose should be to improve strategic stability -- in simplest terms, to reduce the risk of nuclear war. Yet the administration seems committed to measures in the fast-moving strategic arms negotiations that could decrease stability and damage our allies' confidence in our deterrent.

Two such measures are of special concern. In the context of an agreement providing a 50 percent cut in strategic warheads -- down to a level of 6,000 -- the administration has been striving to ban mobile ICBMs. Perhaps as a consequence it has been, at best, only weakly supporting its own small mobile ICBM. It has also apparently decided to test 12, rather than eight, warheads on the new Trident II submarine-launched missile.

Taken together, these two policies will have dramatic consequences. As we approach the turn of the century our only ICBMs will be located at fixed, easily targetable positions. Furthermore, each of our ballistic missile submarines will be counted, under any realistic arms control scenario, as carrying nearly 300 warheads. In the strategic agreements now being discussed, around 1,200 of 6,000 warheads will be reserved for bombers and their weapons; negotiations will determine the size of each side's ballistic missile forces, but even if we retain only 50 MX and about 200 Minuteman III missiles, there could not be more than around 3,600 warheads available for U.S. submarines. Given the counting rules that the administration seems to want to ignore, we would not be permitted more than a dozen ballistic missile submarines, of which only eight or so would normally be at sea.

Under such a prospective agreement, the Soviets would still have ample warheads to allocate to the destruction of our fixed ICBMs. In such circumstances, eight or so submarines are very few baskets in which to put the nation's entire survivable strategic nuclear deterrent. This is especially alarming when one looks at a Soviet force of well over 100 nuclear attack submarines that could threaten this handful of Tridents.

It will be said in response that our ICBMs and bombers could survive a Soviet attack -- that we needn't rely entirely on submarines. But given the march of technology and the course the administration has set, the 1990s will bring serious vulnerabilities for the bombers on their bases and for nonmobile ICBMs. Today, the Soviets have a hard time coordinating an attack on these two land-based parts of our strategic forces because of the different flight times of their ICBMs attacking our ICBM silos (half an hour) and of their submarine-launched missiles attacking our bombers (a few minutes). In the not-too-distant future, however, that coordination problem will disappear for Soviet planners as they become able in a surprise attack to destroy both our bombers on their bases and our ICBMs in their silos (as well as "rail-garrison" MX in its roundhouses) by firing accurate MIRVed missiles with very short flight times from submarines pulled close to our shores.

Thus before too many years our MX and Minuteman ICBMs, even taken together with our bombers, will be vulnerable in their current and proposed basing. It seemed, until recently, that the nation's efforts had been directed toward solving this problem, toward having a survivable ICBM, by employing mobility. Mobility is the one assured way of having a survivable ICBM force in such an environment. But in spite of the threat, the administration is busily abandoning the mobile ICBM, both by its effort to ban mobiles in a strategic arms agreement and by its willingness to let funding for its own small mobile ICBM program be undermined by some of its own officials and by its friends in Congress.

The result of these developments will be, in the relatively near future, a vulnerable land-based ICBM and bomber force and only a few submarines to carry our whole survivable strategic deterrent. In these circumstances we are driven toward one of two choices: decide now to bet that we will be able to deploy survivable, effective, affordable defenses against ballistic and cruise missiles, or that we will adopt the policy of launching ICBMs on warning alone. The first approach is fraught with technical uncertainty; the second risks accidental nuclear war in case of a false alarm and has been resisted for decades by all thoughtful political and military experts.

There is some dissatisfaction in the Senate with the forthcoming INF treaty. But for those concerned about deterrence and strategic stability, that treaty is the wrong target. The INF accord has been handled in such a way as to create substantial political problems for NATO, but with the right follow-on approach to NATO modernization and arms control, such problems can be satisfactorily resolved. The INF treaty does, above all, lead us to understand the need to consult closely with our allies long before committing ourselves to an agreement, as well as the danger of making proposals whose principal merit is the expectation that the Soviets will reject them.

But the downside risk of the continuing strategic negotiations is far greater than that of the INF treaty. Without changes in the two current administration policies that are creating the survivability problem, the United States could well be on the verge of committing itself to a strategic treaty that many moderate members of Congress and thoughtful citizens will reluctantly conclude is clearly contrary to the national interest.

Brent Scowcroft was national security adviser, 1975-1977. John Deutch was undersecretary of energy, 1977-1979. R. James Woolsey was undersecretary of the Navy, 1977-1979.