Lisbeth Gronlund

One of the major foreign policy issues you will face early in your presidency is whether and how to proceed with national missile defense (NMD). Given the technical problems and resulting delays in the current program, the issue for at least your first term is not whether and how to deploy, but whether and how to continue R&D, and whether and how to proceed on the political front with other countries.

It's not clear the United States will ever need NMD to deal with emerging missile states; after all, there's no long-range threat yet, and if one emerges, there's no reason to believe that deterrence won't work. And NMD is not the best way -- or even a good way -- to deal with the problem of accidental or unauthorized launches from Russia.

Moreover, based on my analysis of the planned mid-course NMD, deploying that system would almost certainly be a net negative for U.S. security. It is likely to be ineffective against a real attack, should one occur; U.S. planners will have no basis for confidence in the system's effectiveness; and its deployment could well entail large security costs in the form of negative reactions from Russia and China.

Should you nonetheless decide to proceed on some level with the planned NMD -- or some other system -- here's how to do so:

Understand the technical realities that the NMD program faces, and help explain them to the public.

Get the straight scoop from your military and scientific advisers, not the PR talk. And then level with the public. Misrepresenting the program is a disservice to everyone, including proponents of the system, and it will only serve to undercut U.S. security. The program should be based on sound science, not wishful thinking.

The reality is that the system as planned has a long way to go before it can even "walk" -- that is, reliably intercept a mock warhead on the test range. Only one of three intercept tests has succeeded in hitting the target.

More important, the system may never be able to "run." The real challenge in fielding an effective national missile defense is to be able to intercept real-world warheads, namely those that incorporate "countermeasures" designed to confuse or overwhelm the defense. Even attempting to demonstrate that capability isn't in the cards for several years. The Pentagon isn't planning to do any intercept tests that would require the system to perform against realistic countermeasures through at least your first term in office.

Revamp the nmd test program so it can tell you what you need to know.

Only targets that correspond to what the Pentagon has termed the "defined C-1 threat" are scheduled to be used in the intercept tests. But should an emerging missile state develop and deploy long-range ballistic missiles, their warheads and decoys are unlikely to resemble the Pentagon's "defined" threat.

In fact, the targets the NMD system will be tested against exclude the very countermeasures that will be available to new missile states, according to the U.S. intelligence community. Ballistic Missile Defense Organization Director Lt. Gen. Ronald Kadish told the Senate Armed Services Committee last June that the test program will not include "many" of the countermeasure technologies that U.S. intelligence agencies identified as being "readily available."

If you continue the test program with this definition of threat, the tests will tell you nothing about the ability of the proposed system to intercept real-world targets -- which is information you need to make a sound decision about continued development or deployment.

A revamped and strengthened test program should have the strong support of both congressional Democrats and Republicans -- after all, who wants a defense that won't work? You should do three things to make sure that adequate testing takes place:

• Advocate for an increased nmd test budget.
  This is not the place to scrimp. If the test program is to establish that the NMD system can reliably intercept real-world targets, then many more tests -- and more realistic tests -- will be needed than are planned through the next several years.

• Insure independent review of test design and results.
  Philip Coyle, the Pentagon's director of testing and evaluation, has done an admirable job of assessing the NMD test program in a fair and nonpartisan manner and providing sound advice to the secretary of defense and Congress. You should keep Coyle in this important position or appoint a new director who is an equally independent straight-shooter. But because the director of testing and evaluation has no authority and serves in an advisory capacity, his advice can be ignored by advocates of new weapons systems. To remedy this problem, you should set up a standing high-level independent review panel to oversee the test program.

• Set up an independent countermeasures "Red Team"
  to develop, build, and test countermeasures using technology available to emerging missile states. It is clear that the planned NMD system will be highly vulnerable to at least some types of countermeasures. The key question is: What kind of countermeasures will emerging missile states be able to develop and deploy?
  
  The lack of intelligence information on the countermeasure programs of emerging missile states requires that you answer this question in a rigorous way -- by using a Red Team to assess what countermeasures a specific country is capable of developing. To avoid a conflict of interest, it is essential that this team not be under the financial control or oversight of the Ballistic Missile Defense Organization. The planned NMD system should then be tested against the countermeasures the Red Team determines would be available to potential attackers.

Fly before you buy.

The United States makes sure all its new weapons systems are tested before it decides to deploy -- and NMD should be no exception. You should only consider deploying a system once the test program has demonstrated that it can reliably intercept real-world targets using countermeasures.

Slow down.

Return to a research and development program rather than continue the "develop for deployment" program. And, as an alternative to the planned system, which is designed to destroy targets in mid-course, you should consider whether to begin R&D on a boost-phase system, which might be more effective and carry fewer strategic costs.

Some NMD supporters will criticize you for returning to an R&D program, but you can explain that it is the only prudent step to take given the technical realities. You should also point out that if you continue on the current path, even the first phase of the system would not be operational before at least 2007, and probably not before the end of your second term.

Thus, there is no reason to begin deploying any components, including the X-band radar in Alaska, for several years. Better to use the time to see if the technology is up to snuff, and to make progress on the political front with Russia and other countries.

Pursue diplomatic approaches to dealing with emerging missile states.

Doing so is not idealistic, but extremely pragmatic: To the extent that the North Korean missile program poses a security threat to the United States, the only feasible near-term solution to the problem is political, not technical. Continue President Bill Clinton's efforts to engage North Korea and to make permanent that country's suspension of its long-range missile program. Give your full attention and commitment to political solutions, and don't short-change this approach.

Finally, and most important, look at the big picture.

Move forward with deployment of an NMD system only if it is a net plus to overall U.S. security. You have to consider the reactions of other countries to NMD deployment, especially Russia and China. These countries do not -- and should not -- hold a veto over U.S. decisions, but their reactions will affect U.S. security more than anything North Korea does.

It also means you have to get your priorities straight. The biggest nuclear danger to U.S. security is an accidental, unauthorized, or erroneous Russian launch, and your first priority should be to deal with that threat. Perversely, NMD deployment could prevent you from solving the problem -- because to be confident it can still retaliate after a U.S. attack, Russia is likely to respond to deployment by increasing its reliance on launch-on-warning.

The evolving relationship with China will also be affected, as will the pace and scope of China's now-quite-modest nuclear modernization program. Spurring China to build up its nuclear forces to overwhelm a U.S. NMD could in turn spur India and then Pakistan to build up their forces. Such developments are certainly not in U.S. interests.

Also, to the extent you are worried about emerging states, keep in mind that the United States needs Russian and Chinese cooperation to control missile and nuclear technology transfers to other countries.

You will need to have serious and sustained discussions with Russia and China. These talks must be a two-way dialogue; if you expect them to take U.S. security concerns into account, you must take their concerns into account as well. You should commit your administration to getting Russian agreement on any ABM Treaty changes, and you should not pull out of the treaty if you don't get quick agreement. Given the state of technology, there is no urgency -- you have a lot of time to try to work it out.

Lisbeth Gronlund, a member of the Bulletin's Board of Directors, is a senior staff scientist at the Union of Concerned Scientists and a research fellow in the Security Studies Program at MIT.