When George Bush and Mikhail Gorbachev meet this week, they will have a chance to improve -- quickly and substantially -- the Strategic Arms Reduction Treaty. But they can only do this if they have the vision to rise above the trivial criticisms currently being leveled at START from both the right and the left, and to focus instead on correcting the treaty's genuine shortcoming: its failure to promote stability.

Here at home, START is under attack by the political right for allegedly giving away the store to the Soviets. This is nothing new. Anyone can take any agreement, vocalize over the provisions that favor the other side and pretend those favoring his own side don't exist. Over the years, American right-wingers have developed this nonsense technique to a fine art.

Now with the advent of glasnost, Soviet right-wingers appear to be doing the same thing in mirror image. In the Soviet Union, START is under attack for giving away the store to the Americans.

Both sets of conservative critics are wrong and should be disregarded. The treaty is well-balanced. But just because it restrains both parties equally doesn't necessarily make it good arms control. This is the basis of the attack START is receiving from the left.

Left-wingers fault START for reducing today's 24,000 superpower strategic nuclear warheads by a much smaller factor than the 50 percent originally advertised. But they, too, miss the mark. Even a real 50 percent reduction would be a mere cosmetic exercise. The difference between the world after a nuclear war using 24,000 warheads and the world after a 12,000-warhead war would not be readily apparent. The only tolerable nuclear war is one that never begins.

That's what arms control is really about: preventing wars from beginning. It can do this by creating stability -- that is, making a Pearl Harbor-type nuclear surprise attack infeasible. Simply reducing warheads doesn't in itself aid stability. Instead, stability requires controls on the nuclear weapons properties most suited to striking the other side first and preventing retaliation. These properties include accuracy, reliability, warhead multiplicity and potential for surprise.

START should control these first-strike properties. Unfortunately, it doesn't.

This is its significant failure.

START takes only one token step toward stability: by shifting some warheads from quick-striking ballistic missiles to slow-striking bombers and cruise missiles, it theoretically reduces a nuclear aggressor's ability to surprise. But since START leaves 9,800 quick ballistic missile warheads, the shift is symbolic where it needs to be real.

I support START, but it needs to be better. It needs to lock in stability now, because the immediate next generation of weapons could be much quicker and more destabilizing than any addressed by START. If it were to head off these new surprise-attack weapons, START could make a genuine contribution to stability.

The two most destabilizing near-term potential devices are anti-satellite weapons and short-time-of-flight ballistic missiles. An aggressor planning a nuclear Pearl Harbor could use either or both devices for a surprise attack not achievable by other means.

1) Anti-satellite (ASAT) weapons. Early warning satellites are the only means of detecting hostile intercontinental ballistic missiles at the beginning of their 30-minute flight. If hostile ASATs destroyed these satellites, the targeted nation would be unable to see approaching ICBMs until they came over the horizon. Today's 30-minute warning time would be cut in half.

This 15-minute difference doesn't sound like much, but it could grievously undermine stability. With today's 30 minutes' warning we can assess an incoming attack, notify the president and give him time to decide whether to launch retaliatory IBCMs out of their silos. If an aggressor could cut our warning time to 15 minutes, we'd be unable to do these things. The president would be unable to retaliate with his ICBMs.

2) Short-time-of-flight (STOF) ballistic missiles. Flying low and fast after being launched from submarines close offshore, specially designed STOF missiles could strike inland targets in about five minutes -- more than twice as quickly as any other strategic weapon. While STOF missiles would lack the accuracy to destroy hard ICBM silos, they don't need accuracy to do a devastating preemptive strike against manned bombers on the ground. Bombers' survival depends on being able to flush safely airborne, which takes about eight minutes. Today, they have time to do that. In a STOF world, they wouldn't. We would be unable to retaliate with our bombers.

Two of the three branches of our nuclear retaliatory deterrent would be useless. While we would retain the nuclear deterrent in our submarines, placing all our eggs in one basket would create a world far less stable than that of today. Both ASAT weapons and STOF missiles would tighten the nuclear hair trigger, weaken deterrence of aggression and undermine stability.

Neither device now exists in any effective form on either side, but both are achievable with current technology. Both sides could have both weapons after a few years of development and flight testing. But flight testing, which experience shows to be absolutely necessary for high-confidence deployment, can be prevented. Given properly negotiated and verifiable testing prohibitions, neither side could get militarily usable ASATs or STOF missiles. These weapons are ideal material for arms control.

Each side is ready to solve half the problem. The Soviets have wisely proposed to ban flight testing of ASATs and to dismantle the ancient and ineffective ASATs that now exist. Similarly, the Bush administration has wisely proposed to ban STOF flight tests.

Unfortunately, neither side has responded to the other's offer in any constructive way. The two half-full glasses remain half empty to history.

The two nations need to combine the best features of their respective negotiating positions. In both cases, Bush and Gorbachev should just say yes.

The writer, a Democratic representative from Oregon, is an official congressional adviser to the U.S. START delegation.