Sometime in the morning of March 12, U.S. and Soviet negotiating teams will sit down at a table in a converted office building overlooking Lake Geneva. The chief of the U.S. Strategic Arms Reduction Treaty (START) delegation will be John Tower, who in his long service in the Senate has built a formidable reputation as an opponent of SALT II, the nuclear explosion treaties, and arms control in general.

He will be well aware that his major mandate has little to do with sitting at that table and hammering out the control of strategic arms. He will be bringing every ounce of concentration to bear on the task at hand: telephoning senators and congressmen in Washington to solicit their support for the MX.

"Don't tie my hands; give me the MX missi I can get arms reductions," Tower will tell us. Some will be tempted to agree, as did a recent editorial in The Post. In our hunger for peace, some of us may wish to equate the simple act of sitting at a table with the aggressive and capable pursuit of security through arms control.

But does the administration's record give us reason to believe it can conclude an agreement, if only we will pay for it by building a vulnerable weapon? After four years in office -- including more than three in which the MX was funded -- the administration is, at best, back where it started.

More important, are the administration's proposals anything worth paying for? For several years, the core of Reagan's START plan has been (1) to limit each side to 5,000 warheads deployed on ICBM and submarine- launched ballistic missiles (SLBMs) combined, and (2) to permit these warheads to be replaced, improved and modernized in any way.

But our representatives will bear to Geneva nothing of national security benefit. Tower and his colleagues are advocating no proposition of sufficient merit to justify tens of billions for an otherwise worthless, vulnerable MX missile.

Suppose the Soviet team were to stun us all by instantly accepting the full American arms control position. Would we be better off? No. The purpose of arms control is to reduce the danger of nuclear war -- to strengthen deterrence by reducing Soviet first-strike capability significantly below what it would otherwise be. But under the Reagan national security program -- weapons and arms control combined -- in the next decade the Soviet first-strike threat will mushroom:

Pinpoint accuracy will enable them to destroy our hardened ICBM silos and command centers.

Paralyzing surprise will enable them to destroy our ICBMs and bombers before they can be launched safely. Such surprise will come from fast, submarine-launched ballistic missiles fired close off our coasts.

Three or more quick, accurate Soviet warheads for each U.S. ICBM silo will enable the other side to drop a lethal warhead on each of our 1,000 silos, follow with a second warhead and still have a massive force remaining to destroy the United States.

The Soviets do not yet have this deadly combination. But this will change to the aggressor's advantage if Reagan's plans, including MX and START, run their course.

The U.S. Air Force has released figures showing that, within a few years, ever-improving Soviet ICBM accuracy will reduce our silo survivability to near zero. Even more disturbing are the Soviet advances we should expect to follow shortly thereafter: Satellite-guided SLBMs accurate enough to render further U.S. silo hardening irrelevant, and quick enough to destroy both our ICBMs and bombers before they can be launched; maneuvering warheads which could be guided directly to any small mobile ICBMs we might deploy to escape our vulnerable silos.

At this point, there is no prudence in building new ICBMs and bombers that would work better than the present models if somehow they survived attack, when we know they will not survive. There is no sense in building the MX and putting it in vulnerable silos where, as John Tower himself has said, it will just "give the Russians more sitting ducks to shoot at." If it were not for the seductive mirage of an arms control connection, we would have shut down the MX long ago.

We can do better. An effective and verifiable solution has been endorsed by the House of Representatives: Ban the flight testing of ballistic missiles as part of the nuclear freeze or as a free-standing treaty. Untestable missiles can never be made quicker, more accurate or more maneuverable. With Soviet accuracy thus frozen, we could make our present ICBMs highly survivable by superhardening their silos. A flight test ban would close the ICBM "window of vulnerability" that will otherwise continue to open under the Reagan program.

Reagan will have no part of a flight test ban because it would destroy our own plans for new missiles. So under his program we will build modern, vulnerable missiles that will only be good for shooting first, and the Soviets will do the same -- partly on their own incentive, partly in reaction to ours.

In the space weapons negotiations, the secretary of defense has explicitly ruled out controls on testing of weapons under development and on procurement of weapons that are found to be workable. Apparently we will only seek to have the Soviets give up weapons that don't work.

Geneva may appear to some to be about arms control; for my part, I can find noth there but a rhetorical fog under which lies no plan that will either reduce or control arms. This administration, its concern for military security notwithstanding, proposes nothing that would halt the onrush of Soviet strategic first-strike capability because it covets these weapons itself.

When Tower asks us not to tie his hands, perhaps we should point out that, when all you're doing is twiddling your thumbs, tied hands are no handicap.